Clinical review of non-invasive ventilation

Abstract

Non-invasive ventilation (NIV) is the mainstay to treat patients who need augmentation of ventilation for acute and chronic forms of respiratory failure. The last several decades have witnessed an extension of the indications for NIV to a variety of acute and chronic lung diseases. Evolving advancements in technology and personalised approaches to patient care make it feasible to prioritise patient-centred care models that deliver home-based management using telemonitoring and telemedicine systems support. These trends may improve patient outcomes, reduce healthcare costs and improve the quality of life for patients who suffer from chronic diseases that precipitate respiratory failure.

Shareable abstract

Non-invasive ventilation (NIV) has become the mainstay for treatment of patients who need augmentation of ventilation to treat acute and chronic forms of respiratory failure. Herein, we provide a succinct review of NIV. https://bit.ly/3AwFC4p

Introduction

Non-invasive ventilation (NIV) has become the mainstay treatment for patients with respiratory failure. NIV can be delivered via various modes and titrated to meet a patient's specific respiratory demand. It avoids the risks associated with intubation and invasive mechanical ventilation (IMV) while simultaneously improving patient comfort and mobility. NIV can be initiated rapidly and discontinued quickly, making it invaluable to manage acute or chronic forms of respiratory failure. As a result, NIV has been utilised in many different respiratory diseases and clinical scenarios using different forms of interfaces and delivery systems to maximise patient comfort as well as achieve satisfactory ventilation. Herein, we provide a comprehensive review of NIV.

Physiological and clinical benefits of NIV

NIV refers to mechanical ventilation without an artificial airway. NIV, in its most basic form, uses a mask to deliver gas from a positive pressure ventilator to the patient's mouth and nose. The physiological benefits of NIV are detailed in figure 1.

FIGURE 1.

Physiological benefits of non-invasive ventilation (NIV). There are numerous physiological benefits of NIV, as detailed here. In addition to pulmonary effects, NIV causes physiological and clinical improvements in the cardiovascular and neurological systems. CPAP: continuous positive airway pressure; FRC: functional residual capacity; IMV: invasive mechanical ventilation; LV: left ventricular; PEEP: positive end-expiratory pressure; V/Q: ventilation/perfusion.

NIV improves respiratory function through several mechanisms. Primarily, it decreases the effort required for breathing by employing supra-atmospheric pressure intermittently to the airways, which elevates transpulmonary pressure, expands the lungs, increases tidal volume and alleviates strain on the inspiratory muscles. Exhalation occurs passively with lung recoil. NIV also decreases diaphragmatic exertion [1, 2]. In patients with severe COPD exacerbations, the combination of positive end-expiratory pressure (PEEP) with inspiratory pressure support further lessens breathing effort by counteracting the effects of auto-PEEP [3]. This reduces diaphragmatic pressure swings more effectively than either pressure support or PEEP alone, decreasing respiratory rate, muscle activity, breathlessness and carbon dioxide retention. Additionally, NIV increases functional residual capacity, which helps with alveolar recruitment, decreases shunt and improves ventilation/perfusion matching. These effects improve oxygenation and potentially alleviate dyspnoea by shifting the respiratory system to a more compliant region of the pressure–volume curve.

NIV can enhance left ventricular function by reducing afterload through increased intrathoracic pressure. This lowers transmyocardial pressure and improves cardiac output [4].

A major reason to use NIV is to avoid complications incurred by using IMV. These can be categorised as complications related to the insertion of an artificial airway, loss of airway defences and complications that may occur following artificial airway removal [5].

NIV maintains the integrity of the upper airway, safeguards air passage defence mechanisms and permits patients to engage in activities such as eating, drinking, speaking and clearing secretions. Additionally, it improves patient comfort, convenience and mobility, often at a lower cost compared to endotracheal intubation [6, 7].

A significant advantage of NIV, leading to reduced complication rates, mortality and hospital stays, is its ability to decrease healthcare-acquired infections. Studies show a considerable reduction in rates of healthcare-acquired pneumonia compared to physiologically similar patients intubated via endotracheal tubes, even after adjusting for illness severity. Patients receiving NIV undergo fewer invasive procedures, such as urinary catheterisation or central venous lines, likely contributing to a lower incidence of healthcare-associated infections and sepsis [8, 9].

The use of NIV in chronic respiratory failure offers improved ventilation during sleep by preventing upper airway collapse [9]. NIV also improves daytime gas exchange possibly through three distinct mechanisms. 1) NIV may rest chronically fatigued respiratory muscles and improve strength, thereby allowing for more efficient ventilation during the day [10, 11]. 2) By alleviating microatelectasis, NIV improves lung compliance and increases forced vital capacity, thereby facilitating improved gas exchange during periods when NIV is not used [12]. 3) By improving chronic hypercapnia, NIV can improve respiratory function by a gradual excretion of bicarbonate and restore carbon dioxide sensitivity of the respiratory centre to a more normal level, thereby reducing daytime hypoventilation [13]. Through these mechanisms, NIV can improve ventilation in patients with chronic hypoventilation and sleep disordered breathing.

Indications for NIV

Pertinent randomised controlled trials investigating NIV in acute and chronic respiratory failure are detailed in tables 1 and 2, respectively. Characteristics of patients that benefit the most from NIV in these settings are described in table 3.

TABLE 1.

Randomised controlled trials investigating non-invasive ventilation (NIV) in acute respiratory failure

First author, year [ref.]	Design	Population, sample size and intervention	Outcome
AECOPD
Bott, 1993 [16]	Prospective, randomised controlled trial	60 patients with AECOPD and hypercapnia; conventional treatment or conventional treatment plus bilevel NIV	NIV group had improved pH, reduced hypercapnia, reduced dyspnoea and reduced mortality
Brochard, 1995 [2]	Prospective, randomised trial	85 patients with AECOPD and respiratory acidosis; standard treatment or standard treatment plus bilevel NIV via face mask	NIV group had reduced need for IMV, reduced length of stay and reduced in-hospital mortality
Barbé, 1996 [15]	Prospective, randomised controlled trial	24 patients with AECOPD in ER; standard treatment versus standard treatment plus BiPAP	All patients improved; NIV did not show a difference compared to standard treatment in change in oxygenation, respiratory acidosis or airflow obstruction
Celikel, 1998 [17]	Single-centre, prospective, randomised controlled study	30 patients with acute hypercapnic respiratory failure; standard treatment versus standard treatment plus bilevel NIV	NIV group had improved gas exchange, reduced hospital length of stay and reduced need for IMV
Avdeev, 1998 [20]	Prospective, randomised trial	58 patients with AECOPD, hypercapnia; conventional therapy or conventional therapy plus BiPAP	BiPAP improved respiratory acidosis, reduced need for IMV, reduced mortality and shortened hospital length of stay
Plant, 2000 [19]	Prospective, multicentre, randomised controlled trial	236 patients with AECOPD and mild-to-moderate respiratory acidosis; standard treatment or standard treatment plus bilevel NIV	NIV group had improved respiratory acidosis, respiratory rate and breathlessness, and reduced need for intubation and in-hospital mortality
Khilnani, 2010 [18]	Prospective, randomised controlled trial	40 patients with AECOPD and respiratory acidosis; conventional therapy or conventional therapy plus BiPAP	BiPAP group had improvement in respiratory acidosis and reduced need for intubation; in-hospital mortality was not reduced with NIV
Acute exacerbations of asthma
Holley, 2001 [28]	Prospective, randomised controlled trial	35 patients with status asthmaticus; standard initial treatment alone or standard initial treatment plus BiPAP	BiPAP group had non-significant trend towards decreased need for mechanical ventilation and decreased hospital length of stay
Soroksky, 2003 [23]	Prospective, randomised, placebo-controlled study	30 patients with severe asthma exacerbation; conventional therapy or conventional therapy plus BiPAP	BiPAP showed statistically significant improvement in FEV1 and reduction in need for hospitalisation
Gupta, 2010 [24]	Prospective, randomised controlled trial	53 patients with severe acute asthma; standard medical therapy or standard medical therapy plus bilevel NIV	More patients receiving NIV experienced >50% improvement in FEV1 at 1, 2 and 4 h compared to standard therapy, although not statistically significant; significantly shorter ICU stay in NIV group
Soma, 2008 [25]	Prospective, randomised study	44 patients with mild-to-moderate acute asthma exacerbation; standard medical therapy or standard therapy plus low-pressure bilevel NIV or standard therapy plus high-pressure bilevel NIV	Statistically significant improvement to FEV1 in high-pressure group only compared to conventional therapy group; significant improvement to dyspnoea as measured by Borg scale in high- and low-pressure groups
Brandao, 2009 [26]	Prospective, randomised controlled study	36 patients with severe asthma exacerbations; nebulisation with simple face mask or nebulisation with bilevel NIV 15/5 cmH2O or nebulisation with bilevel NIV 15/10 cmH2O	Statistically significant improvement to PEFR and FVC in high EPAP group; statistically significant improvement in PEFR and respiratory rate in low EPAP group
Galindo-Filho, 2013 [27]	Randomised, controlled trial	21 adults with moderate-to-severe asthma exacerbation; nebulisation alone or nebulisation plus NIV	Statistically significant improvement in respiratory rate, FEV1, FVC, PEFR and inspiratory capacity in NIV group
Acute heart failure
Masip, 2000 [31]	Prospective, randomised trial	40 patients with acute cardiogenic pulmonary oedema; conventional oxygen therapy or bilevel NIV	NIV group had shorter resolution time of symptoms and rapid improvement in oxygenation; there was no difference in hospital length of stay or mortality
Nava, 2003 [32]	Prospective, multicentre, randomised trial	130 patients with acute respiratory failure due to cardiogenic pulmonary oedema; medical therapy plus supplemental oxygen or medical therapy plus bilevel NIV	NIV group had faster improvement in oxygenation and symptoms; overall, need for intubation, hospital mortality and length of stay were similar; in subgroup of hypercapnic patients, NIV led to reduced need for intubation
Park, 2004 [33]	Prospective, randomised trial	80 patients with severe acute cardiogenic pulmonary oedema; oxygen via face mask or CPAP or BiPAP	Both CPAP and BiPAP led to significant improvement in oxygenation, dyspnoea, tachypnoea and tachycardia along with reduced intubation and mortality at 15 days; overall, in-hospital mortality was not different
Gray, 2008 [30]	Multicentre, prospective, randomised controlled trial	1069 patients with acute cardiogenic pulmonary oedema; oxygen therapy or CPAP or bilevel NIV	Use of NIV compared to oxygen therapy led to more rapid improvement in dyspnoea, respiratory acidosis and tachycardia; no difference in mortality or need for intubation
Post-extubation
Jiang, 1999 [39]	Prospective, randomised trial	93 extubated patients; oxygen therapy or BiPAP after extubation	Indiscriminate use of BiPAP after extubation did not reduce need for re-intubation compared to oxygen therapy
Ferrer, 2006 [36]	Prospective, randomised controlled trial	162 patients at increased risk for post-extubation respiratory failure; oxygen therapy or BiPAP for 24 h after extubation	NIV reduced post-extubation respiratory failure and ICU mortality; mortality benefit driven by hypercapnic patients; 90-day survival was improved with NIV in those with hypercapnia
Ferrer, 2009 [37]	Prospective, randomised trial	106 intubated patients with chronic respiratory disorders and hypercapnia; oxygen therapy or NIV for 24 h after extubation	NIV demonstrated reduced respiratory failure after extubation and lower 90-day mortality
Ornico, 2013 [38]	Prospective, randomised, controlled trial	40 patients with COPD and intubated for acute respiratory failure; extubated to oxygen mask or BiPAP	BiPAP after extubation compared to oxygen therapy demonstrated reduced rate of re-intubation and reduced in-hospital mortality
Keenan, 2002 [40]	Randomised, controlled, unblinded study	81 patients with cardiac or respiratory disease or intubated >2 days who developed respiratory failure within 48 h after extubation; oxygen therapy or BiPAP to treat post-extubation respiratory failure	No difference in rate of re-intubation, hospital mortality, duration of mechanical ventilation or hospital length of stay between NIV or oxygen therapy
Esteban, 2004 [41]	Multicentre, prospective, randomised trial	221 patients with respiratory failure within 48 h after extubation; standard medical therapy or NIV to treat respiratory failure	NIV did not reduce need for re-intubation compared to standard therapy; ICU mortality rate was higher in NIV group, with a longer time to re-intubation

AECOPD: acute exacerbation of COPD; ER: emergency room; BiPAP: bilevel positive airway pressure; IMV: invasive mechanical ventilation; FEV₁: forced expiratory volume in 1 s; ICU: intensive care unit; PEFR: peak expiratory flow rate; FVC: forced vital capacity; EPAP: expiratory positive airway pressure; CPAP: continuous positive airway pressure.

TABLE 2.

Randomised controlled trials investigating non-invasive ventilation (NIV) in chronic respiratory failure

First author, year [ref.]	Design	Population, sample size and intervention	Outcome
COPD
Xiang, 2007 [53]	Prospective, randomised trial	40 patients with stable severe COPD after hospital discharge; conventional therapy or conventional therapy and bilevel NIV	Long-term NIV use showed significantly improved gas exchange, 6MWD and hospitalisation rates; mortality was similar
McEvoy, 2009 [52]	Multicentre, prospective, randomised controlled trial	144 patients with severe hypercapnic COPD; LTOT or LTOT plus nocturnal bilevel NIV	Nocturnal bilevel NIV showed an improvement in survival but similar gas exchange; self-reported quality of life was worse in NIV group
Cheung, 2010 [54]	Prospective, randomised controlled trial	47 patients with COPD and recent episode of acute hypercapnic respiratory failure; home bilevel NIV or home CPAP	Continuation of home NIV in patients treated with NIV during a hospitalisation for acute hypercapnic respiratory failure was associated with lower risk of recurrent severe AECOPD when compared to CPAP
Köhnlein, 2014 [51]	Multicentre, prospective, randomised controlled trial	195 patients with GOLD stage IV COPD and chronic hypercapnia; standard treatment or standard treatment plus NIV	Use of NIV targeted to reduce hypercapnia significantly reduced 1-year mortality compared to control
Struik, 2014 [49]	Randomised controlled trial	201 patients with COPD admitted with acute hypercapnic respiratory failure and persistent hypercapnia (>48 h) after termination of ventilatory support; standard treatment or discharged with nocturnal BiPAP	Those discharged with nocturnal BiPAP had similar rates of 1-year readmission for respiratory failure or death compared to standard treatment; NIV group did have improved daytime hypercapnia and trend towards improved quality of life
Murphy, 2017 [55]	Prospective, randomised controlled trial	116 patients with COPD and persistent hypercapnia 2–4 weeks after AECOPD; home oxygen alone or home oxygen plus bilevel NIV	In those with hypercapnia that persisted 2–4 weeks after hospitalisation for AECOPD, time to readmission or death within 12 months was significantly prolonged in NIV group
Pulmonary rehabilitation
Bianchi, 2002 [59]	Prospective, randomised trial	33 patients with stable COPD not on chronic ventilatory support; mask proportional assist ventilation during exercise or spontaneous breathing during exercise	In patients with COPD without chronic hypercapnia, the addition of proportional assist ventilation during exercise training did not improve exercise tolerance, dyspnoea or leg fatigue
Duiverman, 2011 [62]	Prospective, randomised trial	66 patients with COPD and chronic hypercapnic respiratory failure; pulmonary rehabilitation or pulmonary rehabilitation plus nocturnal NIV	NIV with pulmonary rehabilitation improved quality of life, dyspnoea, daytime hypercapnia, oxygenation, 6MWD and FEV1 compared to pulmonary rehabilitation alone
Márquez-Martín, 2014 [61]	Prospective, randomised controlled trial	45 patients with severe COPD; exercise training alone, bilevel NIV alone or exercise training plus bilevel NIV	The combination of bilevel NIV and exercise training improved exercise capacity, inflammatory markers and blood gas values
Gloeckl, 2019 [60]	Prospective, randomised controlled trial	20 patients with COPD and chronic hypercapnic respiratory failure; cycle with high-pressure bilevel NIV and oxygen or cycle with oxygen alone	The addition of NIV as an add-on to oxygen supplementation during exercise training led to increased cycle endurance, improved exercise-induced hypercapnia and improved exertional dyspnoea compared to oxygen alone
OSA
Monasterio, 2001 [176]	Prospective, randomised trial	142 patients with mild sleep apnoea–hypopnoea; sleep hygiene and weight loss or sleep hygiene, weight loss and CPAP	CPAP can be considered in treating patients with mild OSA based on an improvement in symptoms
Barnes, 2002 [177]	Prospective, randomised controlled trial	42 patients with mild OSA; placebo tablet or CPAP	CPAP improved symptoms of OSA, including, snoring, restless sleep, daytime sleepiness and irritability
Woodson, 2003 [178]	Prospective, randomised trial	60 patients with mild-to-moderate OSA; placebo or CPAP or temperature-controlled radiofrequency ablation	CPAP improved quality of life and sleepiness for mild-to-moderate OSA
Becker, 2003 [179]	Prospective, randomised trial	60 patients with moderate-to-severe OSA; effective CPAP or subtherapeutic CPAP	Effective CPAP leads to a substantial reduction in AHI and arterial blood pressure, which predicts a 37% reduction in coronary heart disease risk and 56% reduction in stroke risk
Barnes, 2004 [180]	Prospective, randomised controlled trial	114 patients with mild-to-moderate OSA; placebo tablet or CPAP or mandibular advancement splint	CPAP is superior to mandibular advancement splint in improving sleep fragmentation and hypoxaemia, daytime sleepiness, quality of life, neurobehavioral function and blood pressure
Ip, 2004 [181]	Prospective, randomised trial	40 patients, 28 with OSA and 12 controls; CPAP or no intervention	Patients with moderate-to-severe OSA have endothelial dysfunction, and treatment with CPAP could reverse it; the effect, however, was dependent on ongoing use
Lam, 2007 [182]	Prospective, randomised trial	101 patients with mild-to-moderate OSA; sleep hygiene or CPAP or oral appliance	CPAP produced the best improvement in physiological, symptomatic and quality of life measures, while the oral appliance was slightly less effective
Nguyen, 2010 [183]	Prospective, randomised trial	35 patients with moderate-to-severe OSA; CPAP or sham	CPAP improves myocardial perfusion reserve, microvascular disease and endothelial dysfunction, which may prevent cardiovascular disease
Phillips, 2011 [184]	Prospective, randomised controlled trial	38 patients with moderate-to-severe OSA; placebo CPAP or therapeutic CPAP	CPAP in severe OSA improves postprandial triglyceride and cholesterol levels, which may reduce the risk of cardiovascular events
Weaver, 2012 [185]	Prospective, randomised trial	223 patients with mild-to-moderate OSA; sham CPAP or active CPAP	CPAP treatment improves the functional outcome of patients with self-reported daytime sleepiness
Hoyos, 2012 [186]	Prospective, randomised trial	65 patients with moderate-to-severe OSA; 12 weeks of real CPAP or sham CPAP, then 12 weeks of real CPAP for all	Reducing visceral adiposity in men with severe OSA is less likely to be achieved without NIV
OHS
Borel, 2012 [187]	Prospective, randomised controlled trial	35 patients with mild OHS; lifestyle counselling or NIV	NIV treatment, although improving sleep and blood gas measurements dramatically, did not change inflammatory, metabolic or cardiovascular markers
Masa, 2015 [188]	Multicentre, prospective, randomised controlled study	221 patients with OHS and severe sleep apnoea; bilevel NIV, CPAP or lifestyle modification	Bilevel NIV and CPAP are more effective than lifestyle modification in improving clinical symptoms and polysomnographic parameters; NIV had better respiratory functional improvements than CPAP
Masa, 2016 [189]	Prospective, randomised trial	86 patients with OHS and non-severe OSA; bilevel NIV or lifestyle modifications	NIV is more effective in improving daytime PaCO2, sleepiness and polysomnographic parameters; NIV reduced healthcare resource utilisation, cardiovascular events and mortality
Lopez-Jimenez, 2016 [190]	Prospective, randomised, controlled trial	221 patients with OHS and severe OSA; NIV or lifestyle modification	NIV improved quality of life, PaCO2, hospital readmissions and mortality
Corral, 2018 [191]	Prospective, randomised, controlled trial	221 patients with OHS; CPAP, bilevel NIV or lifestyle modification	NIV is more effective than CPAP or lifestyle modification alone in improving pulmonary hypertension, left ventricular hypertrophy and functional outcomes

OSA: obstructive sleep apnoea; CPAP: continuous positive airway pressure; OHS: obesity hypoventilation syndrome; P_aCO2: arterial carbon dioxide tension; 6MWD: 6-min walk distance; LTOT: long-term oxygen therapy; AECOPD: acute exacerbation of COPD; GOLD: Global Initiative for Chronic Obstructive Lung Disease; BiPAP: bilevel positive airway pressure; FEV₁: forced expiratory volume in 1 s.

TABLE 3.

Characteristics of patients that benefit most from non-invasive ventilation

Condition	Patient characteristics
AECOPD and acute asthma exacerbations	Acute hypercapnic respiratory failure, increased work of breathing, no impaired consciousness, protection of airway
Acute heart failure	Beneficial in terms of symptomatic and oxygenation improvement, additional benefit in those who are hypercapnic in reducing need for invasive mechanical ventilation
Post-extubation	Increased age, hypercapnia, heart failure and/or elevated APACHE II score
Chronic COPD and pulmonary rehabilitation	GOLD stage IV COPD with chronic hypercapnia 2–4 weeks after AECOPD
OSA	More pronounced continuous positive airway pressure benefit as OSA severity increases

AECOPD: acute exacerbation of COPD; APACHE: Acute Physiology and Chronic Health Evaluation; GOLD: Global Initiative for Chronic Obstructive Lung Disease; OSA: obstructive sleep apnoea.

Acute respiratory failure

A broad range of literature supports the use of NIV as supportive management in acute respiratory failure of a variety of aetiologies (table 1). The most common NIV devices with established indications for acute respiratory failure are bilevel positive airway pressure (BiPAP), or bilevel NIV, and continuous positive airway pressure (CPAP) [14]. We describe specific conditions causing acute respiratory failure in which NIV is indicated.

Acute exacerbations of COPD

In acute exacerbations of COPD (AECOPD), bilevel NIV support is indicated in significant tachypnoea (respiratory rate >20–24 breaths·min⁻¹) and hypercapnic respiratory acidosis with pH <7.35, both to prevent and as an alternative to intubation and IMV [14].

An early, small study by Barbé et al. [15] investigating the use of BiPAP in AECOPD failed to show an improvement in oxygenation, hypercapnia or airflow obstruction. However, numerous studies have shown that bilevel NIV during AECOPD leads to more rapid improvement in symptoms and respiratory acidosis, reduced need for intubation and reduced length of hospital stay [2, 16–20]. Bilevel NIV in AECOPD has demonstrated a reduced risk of mortality compared to conventional supplemental oxygen [21, 22].

Acute exacerbations of asthma

Despite similar pathophysiology between AECOPD and acute exacerbations of asthma, there is a lack of robust evidence demonstrating the benefit of bilevel NIV in acute exacerbations of asthma. Notably, the European Respiratory Society (ERS)/American Thoracic Society (ATS) guidelines do not recommend a trial of NIV during acute exacerbations of asthma [14].

However, data analysing the use of bilevel NIV in acute exacerbations of asthma have demonstrated some benefit, particularly when increased airflow obstruction and resistance are present [23–27]. Additionally, a trend towards decreased rates of intubation and hospitalisation duration has been observed [23, 24, 28].

Acute heart failure

In patients with acute respiratory failure due to heart failure, decreased respiratory system compliance and alveolar flooding occurs due to high capillary pressures. NIV decreases left ventricular afterload, which in turn decreases negative pressure swings generated by respiratory muscles and improves respiratory mechanics [29]. The ERS/ATS guidelines for NIV in acute respiratory failure recommend either BiPAP or CPAP for patients with acute respiratory failure due to cardiogenic pulmonary oedema [14].

The largest randomised clinical trial found that NIV more rapidly reduces respiratory distress and metabolic disturbances than standard oxygen therapy in patients with acute cardiogenic pulmonary oedema; no difference in short-term mortality was observed [30]. Smaller studies have demonstrated reduced need for IMV and improvements in dyspnoea, hypercapnia and oxygenation compared to oxygen therapy alone [31–33]. Subsequent meta-analyses have shown reduced mortality with NIV compared to conventional therapy in patients with acute cardiogenic pulmonary oedema [34, 35].

Post-extubation

The use of bilevel NIV is recommended to prevent post-extubation respiratory failure in high-risk patients [14]. Extubation to bilevel NIV in these patients (table 3) reduced re-intubation rate, post-extubation respiratory failure and intensive care unit (ICU) mortality [36–38]. Importantly, bilevel NIV is not recommended post-extubation in patients who do not meet high-risk criteria, since re-intubation rates are not lower [14, 39]. Studies have demonstrated no benefit of bilevel NIV on re-intubation, mortality or length of stay in those with post-extubation respiratory failure in general, and one study even showed higher ICU mortality in those treated with bilevel NIV [40, 41].

Coronavirus disease 2019

The coronavirus disease 2019 (COVID-19) pandemic was hallmarked by severe respiratory failure with high rates of IMV and mortality. NIV to treat respiratory failure in COVID-19 gained traction during the pandemic. Several studies examined the efficacy of NIV, including nasal high flow (NHF), BiPAP and CPAP, to treat acute respiratory failure from COVID-19. Given the rapid increase in the volume of patients with COVID-19 severe respiratory illness, NIV was frequently used outside of ICUs; >60% of patients were discharged alive without IMV [42]. Mortality was reduced with NHF compared to other forms of NIV; NHF use was associated with a reduction in IMV and mortality [43–45]. It is important to note that delayed intubation and IMV was associated with a higher risk of mortality in patients with COVID-19 [46].

Chronic respiratory failure

Long-term NIV is indicated in several chronic medical conditions (table 2). We describe specific conditions causing chronic respiratory failure in which NIV is indicated.

COPD

Both the ERS and ATS recommend long-term domiciliary bilevel NIV in patients with chronic, stable, hypercapnic COPD [47, 48]. Long-term bilevel NIV in chronic, stable, hypercapnic COPD patients improves hypercapnia, quality of life, dyspnoea, readmission rates and lung function [49, 50]. Bilevel NIV improves survival in specific settings, such as when treating severe hypercapnia (termed high-intensity NIV) [51, 52].

It is important to note that long-term domiciliary bilevel NIV initiation is not recommended immediately after hospitalisation for acute hypercapnic respiratory failure; rather, hypercapnia and need for bilevel NIV should be assessed 2–4 weeks after hospitalisation [47, 48]. Several studies examined the use of BiPAP immediately after a hospitalisation for acute hypercapnic respiratory failure and found no difference in mortality and mixed results for a reduction in exacerbations or prolongation of the time to readmission compared to standard care [49, 53, 54]. The HOT-HMV trial reported that patients with persistent hypercapnia 2–4 weeks after an episode of hypercapnic respiratory failure requiring bilevel NIV treated with long-term bilevel NIV had increased time to readmission or death within 12 months compared to those who did not receive NIV. In that study, patients were randomised to receive chronic bilevel NIV in addition to home oxygen or home oxygen alone, with median NIV settings of inspiratory PAP (IPAP) 24 cmH₂O and expiratory PAP (EPAP) 4 cmH₂O. There was an absolute risk reduction of 17% in risk of readmission or death in the group assigned to home oxygen plus NIV [55].

Pulmonary rehabilitation

Much of the evidence for using NIV during rehabilitation pertains to patients with COPD participating in pulmonary rehabilitation. During exercise, expiratory flow limitation and tachypnoea leads to dynamic hyperinflation and tidal breathing at volumes close to total lung capacity, leading to increased work of breathing [56]. NIV has been shown to have numerous physiological benefits in patients with COPD performing exercise. IPAP decreases work of breathing and facilitates respiratory muscle rest, while extrinsic PEEP or CPAP can decrease the inspiratory threshold load on the inspiratory muscles by counteracting intrinsic PEEP [57, 58].

Judicious use of NIV in appropriately selected patients with COPD is essential to deriving clinical benefit. Using NIV during pulmonary rehabilitation in mild COPD without hypercapnia provides no benefit compared to exercise training alone [59]. However, in patients with COPD and chronic hypercapnia, the addition of bilevel NIV to exercise training and pulmonary rehabilitation improves exercise capacity, exertional dyspnoea and exercise-induced hypercapnia [60, 61]. Additionally, the use of nocturnal bilevel NIV in patients with COPD and chronic hypercapnia undergoing pulmonary rehabilitation improves quality of life, dyspnoea, gas exchange, exercise capacity, 6-min walk distance (6MWD) and forced expiratory volume in 1 s decline compared to pulmonary rehabilitation alone [62].

Obstructive sleep apnoea

Numerous randomised controlled trials demonstrate the efficacy of NIV to reduce the apnoea–hypopnoea index (AHI), improve daytime symptoms and reduce cardiovascular risk factors in patients with obstructive sleep apnoea (OSA) [63]. The ERS, ATS and American Academy of Sleep Medicine recommend CPAP as initial therapy for patients with an AHI ≥15 events·h⁻¹ of sleep [64–66]. Overall, CPAP can reduce obstructive events during sleep, daytime symptoms of sleepiness, the risk of motor vehicle crashes and cardiovascular events, and improve systemic blood pressure and quality of life [67]. However, a mortality benefit is less certain [68]. BiPAP is indicated for patients who are unable to tolerate or inadequately respond to CPAP therapy. BiPAP can effectively reduce AHI and improve symptoms in selected patients, particularly those with coexisting conditions such as obesity or hypoventilation [69]. Adaptive servo-ventilation has been investigated as a treatment option for complex sleep apnoea syndrome and central sleep apnoea with varying results. While some studies have shown benefits in reducing AHI and improving symptoms, previous research raised concerns about its safety, particularly in patients with underlying heart failure [70]. Recent evidence has not reported such mortality disparity [71].

Obesity hypoventilation syndrome

Randomised controlled trials and observational studies have consistently determined the efficacy of NIV, particularly BiPAP, in improving gas exchange, reducing hypercapnia and alleviating symptoms in patients with obesity hypoventilation syndrome (OHS) [72]. OHS is diagnosed in patients with obesity (body mass index >30 kg·m⁻²) when awake alveolar hypoventilation (arterial carbon dioxide tension (P_aCO2) >45 mmHg) cannot be attributed to other causes [73]. Most patients with OHS have associated sleep disordered breathing, typically OSA or sleep-related hypoventilation, requiring treatment with NIV [74]. In patients with OHS who have coexisting severe OSA (AHI ≥30 events·h⁻¹), CPAP is the initial mode of choice [75]. For patients with OHS and sleep-related hypoventilation and patients with acutely decompensated OHS, BiPAP is usually the initial modality. Patients with OHS and OSA who fail or do not tolerate CPAP are also treated with BiPAP [72]. For those who fail or do not tolerate BiPAP, a hybrid mode, average volume-assured pressure support (AVAPS), or, less commonly, volume-cycled ventilation may be an alternative [76, 77]. Overall, NIV is the treatment of choice and has been associated with reduced mortality and hospitalisation rates in this population [78]. Hypoxaemia is common in patients with OHS, especially in those with coexisting OSA. Supplemental oxygen should only be administered when NIV alone is insufficient to eliminate hypoxaemia. Supplemental oxygen alone is insufficient therapy for OHS [79].

Neuromuscular disorders

Respiratory muscle weakness can be frequently encountered among patients with a neuromuscular disorder (NMD). NIV can provide symptomatic relief, improve quality of life and, in some cases, increase survival [80, 81]. NIV is indicated in NMD patients with chronic respiratory failure (reduced functional vital capacity, reduced maximal inspiratory pressure or hypercapnia) [81]. The most common conditions for which NIV is recommended include amyotrophic lateral sclerosis (ALS) and Duchenne muscular dystrophy (DMD). Prior studies have shown that in patients with ALS and respiratory insufficiency, BiPAP improves survival, quality of life and sleep-related symptoms [82]. Moreover, Ishikawa et al. [83] demonstrated that bilevel NIV preserves respiratory function and delays respiratory decline in DMD patients. Additional evidence has also highlighted the importance of early NIV initiation in DMD, showing increased survival and improved quality of life; DMD patients receiving nocturnal ventilation had a significantly higher rate of survival to 25 years of age [84, 85]. Proactive monitoring is critical to optimising outcomes in this patient population.

Contraindications and complications of NIV

The evidence for contraindications and complications of NIV varies depending on the specific condition being treated and the type of NIV being used (table 4). It is essential to carefully assess each patient's clinical status, comorbidities and suitability for NIV, and monitor them closely for complications during treatment.

TABLE 4.

Contraindications and complications of non-invasive ventilation (NIV)

Contraindications#
Facial trauma or abnormalities	Severe facial trauma, facial deformities or anatomical abnormalities that prevent proper mask fit may be contraindications for NIV (helmet interface may be able to be used)
Haemodynamic instability	Unstable cardiovascular status, severe hypotension or shock may be contraindications due to the potential for NIV to exacerbate haemodynamic compromise
Bowel obstruction	NIV may be contraindicated in patients with bowel obstruction or severe abdominal distension due to the risk of increased intra-abdominal pressure
Impaired consciousness	Patients with impaired consciousness or inability to protect their airway may not be suitable candidates for NIV due to the risk of aspiration; bilevel NIV may improve impaired consciousness due to hypercapnia, so a balance must be met
Copious secretions or active vomiting	Patients could be at an increased risk of aspiration
Short-term complications
Skin irritation and pressure sores	Prolonged use of NIV masks can lead to skin irritation, pressure sores and ulceration, particularly over bony prominences
Mask discomfort	Some patients may experience discomfort or claustrophobia with NIV masks, leading to poor compliance and treatment discontinuation
Dry mouth and nasal congestion	NIV can cause dryness of the mouth and nasal passages, as well as nasal congestion, which may result in discomfort and sleep disruption
Gastric distension and aspiration	Inappropriate pressure settings or mask leakages can lead to gastric distension and increase the risk of aspiration, especially in patients with impaired airway protection mechanisms
Pneumothorax	High pressures delivered during NIV, particularly with invasive ventilation modes like BiPAP, can increase the risk of pneumothorax, especially in patients with underlying lung pathology [14, 192]
Long-term complications
Ventilator-associated pneumonia	Poor mask hygiene, inadequate cleaning and improper handling of NIV equipment can increase the risk of ventilator-associated pneumonia
Nasal bridge erosion	Chronic use of nasal masks may lead to erosion of the nasal bridge, particularly in patients with prominent nasal bridges or thin skin [48]

BiPAP: bilevel positive airway pressure. ^#: it is important to note that in clinical scenarios involving palliation and/or patients who do not desire invasive mechanical ventilation, NIV may be the most effective and most advanced respiratory support available; in these cases, the use of NIV when these contraindications are present may be acceptable.

Management and monitoring of NIV

When managing and monitoring NIV use, it is important to remember the indication for NIV. Often, chronic respiratory failure does not resolve, and the use of NIV can be indefinite. In these cases, it is important to monitor for efficacy, tolerance and lack of complications (table 4). For example, bilevel NIV settings should be adjusted to target improvement in hypercapnia in chronic hypercapnic respiratory failure due to COPD or OHS, while CPAP settings can be adjusted based on AHI in patients with OSA. In acute respiratory failure, NIV should act as a temporary supportive measure while the underlying cause of respiratory failure is treated (resolution of acute hypoxaemia and/or hypercapnia, normalisation of work of breathing, etc.). NIV should be withdrawn when this is met, or in the opposite instance when respiratory failure worsens or fails to improve (persistent/worsening hypoxaemia, hypercapnia and work of breathing), or contraindications develop; IMV should be considered. Multidisciplinary care involving respiratory therapists, pulmonologists and critical care specialists is necessary to optimise NIV management and minimise adverse effects.

Types of NIV devices

There are a variety of NIV devices available for use, ranging from classic CPAP and BiPAP to newer modalities such as NHF and the non-invasive open ventilation system (NIOV) system. Illustrations of the various types are shown in figure 2 and comparisons are presented in table 5.

FIGURE 2.

Types of non-invasive ventilation (NIV) interfaces. a) NIV helmet interface (Harol). b) Average volume-assured pressure support (AVAPS) mode (V60; Philips Respironics). c) Nasal high flow (Optiflow; Fisher & Paykel Healthcare). d) Non-invasive open ventilation system (NIOV) (Breath Technologies).

TABLE 5.

Comparison of available non-invasive ventilation devices

	Auto-titrating?	Domiciliary?	Supplemental oxygen integration?	Common uses
CPAP	No	Yes	Yes	OSA
BiPAP	No	Yes	Yes	OSA and OHS; obstructive lung disease; acute respiratory failure
VAPS	Yes – varying levels of PS or SV; additional modes can adjust EPAP	Yes	Yes	Chronic respiratory failure (commonly with nocturnal hypoventilation)
Nasal high flow	No	Clinical trials	Yes	Acute respiratory failure
NIOV	Yes	Yes	Yes	COPD; NMDs; kyphoscoliosis
Negative pressure	No	Yes	No	NMDs; kyphoscoliosis

CPAP: continuous positive airway pressure; BiPAP: bilevel positive airway pressure; OSA: obstructive sleep apnoea; OHS: obesity hypoventilation syndrome; VAPS: volume-assured pressure support; PS: pressure support; SV: servo-ventilation; EPAP: expiratory positive airway pressure; NIOV: non-invasive open ventilation; NMD: neuromuscular disorder.

Continuous positive airway pressure

Prior to more widespread use of bilevel NIV, CPAP devices were studied in a randomised controlled trial for initial stabilisation of hypoxaemic respiratory failure (both in patients with and without cardiogenic pulmonary oedema) [86]. CPAP devices provide inspiratory flow (from 0 to 130 L·min⁻¹), an air–oxygen blend to deliver the desired inspiratory oxygen fraction (F_IO2) and PEEP. The Venturi system used to adjust F_IO2 could partially explain the reduction in effective mean airway pressure delivered with higher F_IO2 needs [86]. In patients with a decreased functional residual capacity (i.e. regional atelectasis), CPAP therapy decreases total pulmonary power during inspiration, improves lung compliance and decreases expiratory flow rate [87]. CPAP intolerance in the aforementioned study was 14% [86], comparable to the rates in studies of bilevel NIV [88].

Bilevel NIV

Bilevel NIV augments inspiratory flow through flow or timing cycling. Achieving optimal pressure support and synchrony is key to achieving clinical benefits. A reasonable approach is to set an initial EPAP of 4–6 cmH₂O with titration (often to 8–12 cmH₂O) to achieve consistent triggering and an IPAP of 8–12 cmH₂O with up-titration for an acceptable tidal volume and respiratory rate. A backup rate (spontaneous/timed (BiPAP S/T)) can be set in cases of hypercapnia. F_IO2 and EPAP are adjusted to maintain saturations to the minimum desired goal (typically 88–92%). The maximum peak inspiratory pressure should be kept ≤30 cmH₂O [88–91]. Settings available to maximise patient–device synchrony and comfort include the inspiratory/expiratory timings, inspiratory and expiratory flow triggers and cycling, and mask interfaces [92–94]. Very fast pressurisation settings (200 cmH₂O·s⁻¹) may decrease diaphragmatic effort at the expense of patient discomfort and increased leakage [95].

The data for higher levels of IPAP are more reassuring. A small crossover open-label trial of stable hypercapnic Global Initiative for Chronic Obstructive Lung Disease (GOLD) D COPD patients compared targeting higher IPAP (average 29 cmH₂O) to lower IPAP settings (average 14.8 cmH₂O) [92]. While higher IPAP expectedly leads to higher leak volume, it also resulted in significant inspiratory tidal volume difference, greater reductions in nocturnal and daytime P_aCO2, and greater treatment compliance and tolerance [92].

Helmet interfaces

PAP therapy is often delivered via a mask interface, usually with silicone or foam cushioning. Helmet interfaces use a soft transparent polyvinyl hood with a seal created by a rigid padded collar around the patient's neck; it is secured usually via armpit braces (figure 2a) [90, 94]. As with traditional masks, helmets can be hooked up to a traditional ventilator or BiPAP device. Higher compliance of the delivery interface also allows for use of lower continuous flow rates (closer to 60 L·min⁻¹) and avoids patient asynchrony when patient-generated flow exceeds the flow provided by the blender [94].

As helmets relieve pressure around the nose and nasolabial folds, pressure injuries and leak are less common, although carbon dioxide rebreathing can become a concern without adequate continuous flow (<40 L·min⁻¹) [94]. Helmet devices have been shown to be at least non-inferior (and in some cases more effective) than standard therapies in improving oxygenation, avoiding endotracheal intubation and improving mortality across a variety of hypoxaemic conditions, including COVID-19 [89, 94]. Conversely, the use of NIV through a helmet is less efficient than a standard face mask in hypercapnic failure, particularly with longer inspiratory and expiratory triggering delays, wasted inspiratory patient effort and longer times of asynchrony [90, 93, 94]. Specific attention should be paid to optimal settings to offset these issues, including use of higher cycle-off thresholds (30–60%), a fast inspiratory rise time of 0.5–1 ms, high pressurisation times (80%) and adequate pressure support of 8–15 cmH₂O [89, 90, 93, 94, 96].

Volume-assured pressure support modes

While bilevel NIV is superior to CPAP in providing ventilatory support, its efficacy also relies on adequate patient effort and lung compliance. AVAPS allows for set tidal volumes to be achieved by adjusting the inspiratory flow (figure 2b). As opposed to bilevel ventilation, the range of pressure support is set (typically 6–16 cmH₂O, to a maximum IPAP of 20–26 cmH₂O) to achieve a desired tidal volume (6–8 mL·kg⁻¹ of ideal body weight) [91, 97, 98]. Small series on the use of AVAPS in acutely hypercapnic COPD patients with metabolic encephalopathy showed improvements in Glasgow Coma Scale, carbon dioxide tension (P_CO2), respiratory rate and exhaled tidal volumes compared to BiPAP S/T, with higher average maximum IPAP in the AVAPS group [97]. In stable hypercapnia, 6 months of AVAPS therapy compared to BiPAP S/T led to statistically, but not clinically, significant reductions in P_aCO2, improvements in arterial oxygen tension (P_aO2) and increases in 6MWD; more relevant were improvements in several physical and mental health metrics [91].

“Intelligent” volume-assured pressure support (iVAPS) is a similar proprietary algorithm to provide assured servo-ventilation by learning the patient's waking breathing pattern on minimal EPAP 4 cmH₂O (to a maximum of 15 cmH₂O) and provide targeted alveolar ventilation with backup respiratory rate while compensating for estimated anatomical dead space [98, 99].

In a small crossover trial of patients with chronic respiratory failure (mostly restrictive disease), iVAPS was found non-inferior to standard normal pressure support in treating nocturnal hypoventilation and oxygenation after an in-lab overnight titration study. Compliance time was greater by 1.3 h and patient preference favoured the iVAPS group [99]. Compared to fixed EPAP settings, patients with chronic respiratory failure and hypoventilation syndromes both with OSA [100] and without sleep apnoea [98, 101] may further benefit from the addition of auto-EPAP capabilities to the iVAPS algorithm by increasing time in deep sleep and reducing arousals overnight, and may reduce daytime P_aCO2 levels and patient comfort score [98, 100], although others have suggested non-inferiority with no significant differences in sleep metrics or transcutaneous carbon dioxide levels [101]. Whether auto-EPAP significantly changes the median delivered EPAP or tidal volume or increases leak is unclear [98, 100, 101].

Nasal high flow

While not traditionally grouped into NIV, NHF devices (figure 2c) generate a high degree of humidified airflow and oxygen delivery, decreasing anatomical dead space, reducing carbon dioxide rebreathing and providing a small PAP effect, leading to a decrease in work of breathing [102]. As such, NHF is a highly effective form of non-invasive respiratory support.

In a cohort of stable severe-to-very severe COPD patients (GOLD C/D), NHF modestly increased mean airway pressure by 0.9–3 mbar (similar to prior studies [103]) as flow rate increased from 20 to 50 L·min⁻¹; comparatively, nasal CPAP and bilevel ventilation increased mean airway pressure by 4.96–8.56 mbar [104]. Additionally, NHF decreased respiratory rate by 4.1–5.8 breaths·min⁻¹ and rapid shallow breathing ratio by 13.6–16.5, both measures improving to greater degrees than when using positive pressure devices. Despite a decrease in absolute minute ventilation with lower flow rates compared to spontaneous breathing, P_CO2 levels improve, likely reflecting the beneficial effect of NHF beyond simply improving minute ventilation [104].

Prior upper airway modelling studies have demonstrated more rapid washout of the nasopharynx with NHF in a linear fashion by 1.8 mL·m⁻² for every 1 L·min⁻¹ increase in flow rate; the anterior nasopharynx expectedly is cleared faster than the posterior nasopharynx, with a clearance half-time of ∼0.08 s at 45 L·min⁻¹ [105]. Another study demonstrated that improvements in P_aO2 and P_CO2 achieved at 30 min are similar to values achieved at 60 min [103], again showing the physiological benefits are rapid. Taken together, clearance of the nasopharynx, which accounts for ∼30% of anatomical dead space, reduces carbon dioxide and oxygen-depleted gas rebreathing and decreases the work of breathing [105]. Nasal prong size does not seem to influence improvements in respiratory mechanics beyond influencing comfort [104], although interestingly, high leakage (while reducing delivered mean airway pressure) can improve P_aCO2 reduction, again likely by providing washout of the nasopharynx [103]. These findings led to the development of asymmetrically sized cannula that improve the metabolic work of breathing compared to standard NHF cannulas without significantly changing ventilation or gas exchange [106].

NHF may serve as a substitute to bilevel NIV for COPD management in both the acute and chronic settings. In a small study of 38 acutely hypercapnic COPD patients (defined as P_aCO2 >45 mmHg with pH ≤7.38), Bräunlich and Wirtz [102] demonstrated a 9.1–14.2 mmHg reduction in P_aCO2 without worsening oxygenation at an average flow rate of 25 L·min⁻¹, with greater treatment effect in patients with moderate hypercapnia (pH <7.35) [102]. NHF has been used with success in avoiding re-intubation for patients with hypercapnic respiratory failure (being non-inferior to NIV) [88] and hypoxaemic respiratory failure.

Non-invasive open ventilation

NIOV is a proprietary pneumatically powered, electronically controlled portable non-invasive open ventilation system which delivers augmented breaths (see earlier; figure 2d). Patient-triggered tidal volumes are delivered during a nasal pillow-style cannula with two Venturi-type air entrainment ports. The NIOV ventilator and console requires an oxygen cylinder. At flow rates between 24 and 60 L·min⁻¹, delivered tidal volumes can range from 50 to 250 mL, although depending on the degree of room air entrainment and underlying pathology, patient-delivered volumes and pressures are variable (upwards of 450 mL) [107–109]. At an F_IO2 of 1.0, delivered F_IO2 can range from 0.36 to 0.45 after room air entrainment [107]. Ventilator inspiratory time can be adjusted from 10% to 40% as well as trigger sensitivity to separately titrate rest and ambulatory support [108, 109].

In laboratory models, tidal volume delivery may exceed set ventilator volumes because of room air entrainment at higher volumes (250 mL) with a peak inspiratory pressure of 1.4–9.5 cmH₂O. COPD patients tend to have lower augmented tidal volume deliveries due to higher airways resistance [107]. In studies of COPD patients, NIOV improved both exercise testing and activities of daily living endurance time by 6.1 min, oxygen saturation by 4–6%, reduced respiratory muscle electromyography activity by 22–46% and improved Borg dyspnoea and fatigue scores compared to supplemental oxygen alone [108, 109]. No major complications were noted in a crossover study of home ambulatory use [108].

Negative pressure ventilation

Until positive pressure ventilation was established during the polio epidemic in the 1950s, negative pressure ventilation (NPV) was the primary mode of assisted ventilation in acute respiratory failure [110]. Although trials are limited, NPV has previously demonstrated efficacy in improving acute respiratory failure due to COPD and NMDs [111–115]. In fact, studies comparing NPV with NIV in patients with COPD and acute respiratory failure with hypercapnia showed that both NPV and NIV were equally effective in avoiding intubation and death [116]. The use of NPV for chronic COPD management has had mixed findings in terms of efficacy in improving hypercapnia and lung function [117, 118]. However, NPV for chronic respiratory failure due to NMD has good efficacy in improving hypercapnia, oxygenation and even survival in small, uncontrolled studies [119–121]. Despite the potential benefits of NPV in both acute and chronic respiratory failure, NIV has become widely used for acute and chronic respiratory failure, limiting the routine use of NPV. NPV, however, can be considered when there are contraindications to NIV via mask ventilation, such as reduced mental status due to hypercapnia, excessive airway secretions, facial deformities and inability to fit the mask [110].

It is important to note that in resource-scarce settings, newer NIV modalities such as NIOV, AVAPS/iVAPS, etc., may not be available, and these areas may be limited to conventional NIV (BiPAP, CPAP, etc.); reassuringly there is an abundance of evidence to the efficacy of conventional NIV.

Cost-effectiveness

The cost-effectiveness balance of NIV may be affected by multiple factors, including the health system structure, payment system, medical staff, resources and incidence of underlying respiratory disorders. In 1995, Criner et al. [122] reported improvement in functional status and gas exchange of patients with respiratory failure receiving NIV; however, this treatment incurred higher costs for the health system that were under-reimbursed at that time in the US diagnosis-related group (DRG) system since NIV did not qualify for the ventilator code. This led to a call for a revision of DRG payment scales to recognise NIV as a specific treatment [122]. As the healthcare community gained greater experience with NIV, the costs of care improved as the work demands placed on the staff lessened [8]. Multiple studies report NIV is associated with shorter lengths of hospital stay and reduced costs for patients with respiratory failure due to acute exacerbations of COPD [123–125].

The transition from IMV to NIV in critically ill patients is cost-effective as well. In 1999, Munshi et al.[126] reported significant direct cost avoidance in patient days and unused ventilator days by utilising NIV post-extubation rather than supplemental oxygen in a trauma ICU.

In the outpatient and pre-hospital settings, NIV cost-effectiveness data is not as strong. Several studies have demonstrated that CPAP in the pre-hospital setting is effective in preventing respiratory failure and mortality but more expensive than routine oxygen therapy [127, 128]. The variation in the incidence of eligible patients for NIV was a key factor in the cost-effectiveness of pre-hospital CPAP [129].

In patients with COPD and persistent hypercapnia, NIV was found to be a cost-effective intervention [130]. Advanced NIV (AVAPS) was compared to no NIV or BiPAP in patients with severe COPD; AVAPS was found to have saved the hospital significant costs over a 90-day period compared to the control groups. From the payer standpoint, 3-year cumulative savings with advanced NIV were USD 326 million versus no NIV and USD 1.04 billion versus respiratory assist devices [131].

With regard to sleep-related breathing disorders, Masa et al. [132] demonstrated CPAP is cost-effective in the treatment of OHS with severe OSA. The cost-effectiveness of outpatient versus inpatient NIV setup for OHS is similar [133]. Patel et al. [134] utilised NIV in resource-limited settings in India and demonstrated a 17.7% improvement in survival of COPD patients when the ICU was not available; it was found to be very cost-effective, with an incremental cost-effectiveness ratio of USD 61 per quality-adjusted life-year.

NIV asynchrony: causes and treatment

NIV tolerance has been associated with NIV synchrony and success [135, 136]. Synchrony is the match between the patient's neural inspiratory and expiratory times and the ventilator's mechanical inspiratory and expiratory times [137]. Patient–ventilator asynchrony (PVA) occurs when there is >20% mismatch between pneumatic and neural timing or when there is a mismatch between the patient's respiratory effort and the delivered ventilator support [137–139]. Patient discomfort and poor patient–ventilator interactions lead to PVA and increased rates of NIV failure [140, 141]. PVA is associated with leaks as well as the underlying disease process [142].

NIV synchrony must be achieved to properly unload the inspiratory muscles [137]. PVA causes worsened gas exchange, wasted respiratory efforts and increased discomfort, and will incur increased need for IMV, increased ICU use, sedation, increased length of stay, secondary respiratory complications, morbidity and mortality [136, 143].

Asynchrony with NIV

Types of asynchrony in NIV include trigger delay, ineffective inspiratory efforts, double triggering, auto-triggering, premature cycling and delayed cycling [137, 139, 141, 142, 144, 145].

Patient-related factors also contribute to PVA. Respiratory distress is associated with anxiety. Asynchrony increases the rate of anxiety and increases the risk of NIV failure [146]. Clinicians must be aware of the many factors affecting PVA: mask fitting, leak, dyspnoea, sleep deprivation, anxiety, thirst, dry mucosa, inability to communicate, lack of control, sense of claustrophobia, underlying lung disease, hyperinflation, hypoxia, encephalopathy, fever and other factors [137, 141, 146, 147]. PVA has been reported to occur anywhere from 24% to 43% [136, 139, 143]. NIV failure has been reported to occur in 15–38% in acute respiratory failure [136, 147, 148].

Proper mask fitting and proper inspiratory pressure will help minimise leaks and decrease PVA and NIV failure [137]. Progressive illness can also present as asynchrony due to failure of NIV to provide proper respiratory support [146]. Early detection of PVA allows for a decreased time to intubation in cases of NIV failure; there are overall worse outcomes when intubation is delayed [146].

Several software algorithms have been developed to allow for more sensitivities and earlier detection of PVA [138, 142, 149]. The Neurosync index and SyncSmart allow for automated analysis [138, 149]. The Neurosync index requires the insertion of an oesophageal probe to detect electric activity of the diaphragm. SyncSmart was shown to reduce inter-rater variability by automatically analysing waveforms of pressure and flow [140].

Interface leaks

The magnitude of leaks is associated with a higher rate of intubation [145]. Leaks are calculated as (inspired tidal volume−expired tidal volume)/inspired tidal volume [145]. Leaks can cause loss of extrinsic PEEP and pressure support, and increase in ventilator auto-triggering, rebreathing exhaled gas and increased PVA. There can also be a decrease in F_IO2 and oxygen saturation and an increase in NIV failure [139]. Leaks also reduce patient tolerance and cause awakenings and sleep fragmentation [150, 151]. Late cycling, a consequence of an extension of ventilator insufflation into neural expiration, is another deleterious consequence of leaks [145]. Trigger asynchrony with NIV may be the result of leaks, and it can produce auto-triggering or ineffective triggers [142].

Humidification

Inflammatory mediators released from dried mucosa increase nasal congestion and resistance, potentially causing a decrease in tidal volume, increased work of breathing, patient discomfort, asynchrony and poor compliance with NIV [139]. Humidification has been shown to reduce nasal airway resistance and increase patient comfort and compliance with NIV [139]. Heated humidification has been shown to decrease ICU stay, intubation and mortality [146].

NIV modes and asynchrony

The function of bilevel ventilators depends on the presence of leak. Some bilevel ventilators allow the user to enter the interface being used, to allow more precise identification of the intentional leak [142]. Other bilevel ventilators allow the user to test the leak port as part of the pre-use procedure. Leak detection algorithms adjust for changes in leak with inspiratory and expiratory pressure changes, as well as changes that may occur breath-to-breath due to fit of the interface. Newer generations of bilevel ventilators use redundant leak estimation algorithms [142].

Bilevel ventilators have a set IPAP and EPAP; the difference between IPAP and EPAP is the level of pressure support or pressure control [142]. With pressure-targeted ventilation, the determinants of inspiratory flow are the inspiratory pressure setting and respiratory mechanics (resistance and compliance). The capacity to achieve and maintain IPAP is decreased when intentional leaks increase, suggesting that flow delivery to the patient may be reduced in the presence of leaks, possibly resulting in flow asynchrony [152].

PSV is a patient-triggered and flow-cycled mode of ventilation, in which the patient controls the start of each breath. The initial airflow is high and rapid to meet the targeted pressure. The expiratory valve opens once the maximum inspiratory flow is reached, terminating the inspiratory cycle [139]. Parameters that can be adjusted with PSV are trigger-on threshold, inspiratory rise time, pressure level and, in some ventilators, the cycling-off airflow threshold [139].

Adaptive support ventilation is a closed-loop system that provides a target minute ventilation by automatically regulating the delivered pressure and respiratory rate while keeping the patient's work of breathing to a minimum [147]. It also has a built-in backup respiratory rate in the event of hypopnoea/apnoea [142].

Neurally adjusted ventilatory assist (NAVA) is an assist mode of ventilation that uses electrical activity of the diaphragm, sensed by a special nasogastric catheter, to trigger and terminate the respiratory cycle. The assistance it provides is proportional to the patient's effort and improves patient–ventilation interaction, minimising patient asynchrony [144].

NAVA trigger is not affected by leaks. It is marginally affected by airway pressure, ventilator system flow, intrinsic PEEP, lung volume and air leaks when compared to PSV [138]. NAVA relies on electrical activity of the diaphragm to best synchronise the patient’s respiratory effort and provide ventilatory support more efficiently, reducing neural drive and effort, avoiding over-assistance, decreasing intrinsic PEEP and minimising wasted efforts [143]. The drawback of NAVA is the potential impracticality of the oesophageal probe [143].

New applications of NIV

Wearable devices

Patients with chronic respiratory failure who need non-invasive home mechanical ventilation (HMV) can become deconditioned because of exercise limitation. Limited activity increases mortality in patients with COPD [153]. Several studies have shown that long-term oxygen therapy using conventional delivery tools during exercise did not improve exercise performance or quality of life [154, 155]. Exercise limitation in patients with COPD is due to multiple pathophysiological factors, including ventilatory limitation due to rising end-expiratory lung volume and tidal volume causing dynamic hyperinflation, and reaching O'Donnell's threshold [156]. Other factors include limitations in gas exchange, muscle fatigue and anxiety [157].

Several studies examining NIV during exercise and pulmonary rehabilitation showed promising results [60, 158]. One major issue that comes up is the portability of these devices in the home environment [159,  160]. Carlin et al. [108] reported use of the NIOV system that consists of a wearable, 0.45-kg ventilator and an open, non-sealing, proprietary nasal pillow interface (figure 2d) in 30 patients with chronic COPD. When compared to oxygen therapy, NIOV use resulted in significant improvements in activities of daily living endurance, oxygenation, dyspnoea, fatigue and comfort [108].

Majorski et al. [161] evaluated the effects of portable NIV in both NIV-naïve and NIV-dependent moderate-to-severe COPD patients using the certified VitaBreath device (Philips Respironics). It is a hand-held device specifically designed to provide positive pressure via a mouthpiece, during or after exercise. The portable NIV device provides two fixed pressure levels: 18 cmH₂O during inspiration and 8 cmH₂O during expiration. The study showed significant improvement in Borg dyspnoea scale and 6MWD when using portable NIV. The device is not presently clinically available. These results suggest that more studies are needed to define the group of patients who may benefit from this therapy, Additionally, technical improvement is required to allow adjusting portable NIV settings.

Telemedicine integration: initiation and follow-up visits

The need for telemedicine integration in caring for patients with chronic respiratory failure on HMV arises from different factors. First, the number of patients who depend on HMV is growing significantly [162, 163]. Second, patients prefer home support of their therapy. Lastly, in-hospital resources for elective chronic care are decreasing, especially during and after the recent COVID-19 pandemic.

Few studies have evaluated at-home initiation of chronic NIV in patients with chronic respiratory diseases. Duiverman et al. [164] randomised 67 hypercapnic COPD patients to home or in-hospital initiation of NIV and found that home NIV initiation done via telemedicine monitoring was non-inferior to in-hospital initiation in reducing daytime P_aCO2 levels and improving health-related quality of life at 6 months. Furthermore, the study showed that home initiation of NIV reduced the cost by 50% when compared to in-hospital initiation. Adherence to therapy was better in the home initiation group [164]. Similar results were reported in the Dutch Homerun trial that randomised 96 patients with mostly NMDs [165].

The evidence for telemonitored follow-up in patients with NMDs on HMV is limited. The use of telehealth is especially helpful in severely disabled and rapidly progressive disease patients such as those with ALS. A retrospective, single-blinded controlled trial evaluated 40 ventilated ALS patients and found a reduction in hospitalisations in patients who were followed up via a telehealth modem device versus those who were assessed during regular office visits [166]. Trucco et al. [167] described telemonitoring of peripheral oxygen saturation, heart rate and ventilation data in 48 young ventilated NMD patients in a 2-year case-controlled trial. The study showed fewer and shorter hospitalisations compared to prior to telemedicine start and to those of controls matched for age, disease and severity.

In patients with COPD, telemonitoring follow-up might help in detecting exacerbations and treating them earlier. In a study by Chatwin et al. [168], 68 patients with chronic lung disease including COPD were randomised for a crossover trial with 6 months of standard best practice clinical care (control group) and 6 months with the addition of telemonitoring. 87% of the studied patients were on home NIV. The authors found no difference in the time to first hospitalisation. In fact, hospitalisation rate at 6 months and home visits increased in the telemonitoring versus the control group [168]. Different results were noted by Vitacca et al. [169], who randomised 240 patients with chronic respiratory illness, including those with COPD on NIV, to receive either traditional care or a 1-year tele-assist programme. The study showed that the tele-assist group had significantly fewer acute exacerbations, predominantly in COPD patients. This was most evident in patients using nocturnal NIV [169].

The primary challenge in developing integration of telemedicine in monitoring patients with chronic respiratory failure who need HMV is that it cannot be “that one plan fits all”. This is in part due to the heterogeneity of the diseases that indicate the need for HMV, differences in socioeconomic status, support systems and accessibility to telemedicine monitoring, as well as the complexity and diversity of telemedicine interventions used [170].

Knowledge gaps

The use of NIV is uncertain in some entities, including acute asthma exacerbations, acute respiratory distress syndrome (ARDS) and NHF in hypercapnic respiratory failure. Table 6 lists studies examining these entities.

TABLE 6.

Knowledge gaps in non-invasive ventilation (NIV) and pertinent studies

Acute asthma exacerbations and survival
Althoff et al., 2020 [171]	Retrospective cohort study of more than 50 000 patients with asthma exacerbation; use of NIV was associated with lower odds of receiving IMV and in-hospital mortality; those who progressed from NIV to IMV more likely had pneumonia and severe sepsis
ARDS
Bellani et al., 2017 [172]	Approximately 15% of the 2813 ARDS patients in the LUNG SAFE study received NIV, across all severities of ARDS; NIV use was independently associated with increased ICU mortality; specifically, ICU mortality was worse in NIV than IMV among severe ARDS (PaO2/FIO2 <150 mmHg)
He et al., 2019 [173]	Multicentre randomised controlled trial of 200 patients with pneumonia-induced mild ARDS comparing NIV to supplemental oxygen; NIV group had higher PaO2/FIO2 ratio but the proportion of those requiring IMV or ICU mortality did not decrease compared to routine supplemental oxygen
Hypercapnic respiratory failure and NHF
Bräunlich et al., 2016  [104]	In 54 patients with hypercapnic COPD, hypercapnia decreased with increasing flow delivered by NHF, likely due to washout of the respiratory tract and functional dead space reduction
Sun et al., 2019 [174]	In 82 patients with COPD and acute hypercapnic respiratory failure, use of NHF did not result in increased treatment failure rates (progression to IMV or switch to other treatment group) or increased mortality compared to NIV
Nagata et al., 2022 [175]	104 patients with chronic daytime hypercapnia due to COPD received either home NHF or LTOT; use of NHF significantly reduced the rate of moderate or severe AECOPD and prolonged the duration without such AECOPD; NHF also significantly improved health-related quality of life scores, SpO2 and pulmonary function

IMV: invasive mechanical ventilation; ARDS: acute respiratory distress syndrome; ICU: intensive care unit; P_aO2: arterial oxygen tension; F_IO2: inspiratory oxygen fraction; NHF: nasal high flow; LTOT: long-term oxygen therapy; AECOPD: acute exacerbation of COPD; S_pO2: peripheral oxygen saturation.

Acute asthma exacerbations

Several trials have demonstrated some physiological and symptomatic benefits to NIV in acute asthma exacerbations (table 1), although these trials were underpowered to determine a mortality benefit. However, the use of NIV in acute exacerbations of asthma is increasing, and a large, multicentre retrospective cohort study found that NIV was associated with lower rates of intubation with IMV and mortality [171].

Acute respiratory distress syndrome

The LUNG SAFE study showed that NIV use in ARDS was independently associated with increased ICU mortality, especially severe ARDS [172]. A subsequent study showed that in mild ARDS, NIV improves oxygenation compared to supplemental oxygen but does not reduce the need for IMV or mortality [173]. While there may be a physiological benefit to NIV in mild ARDS, its clinical impact is uncertain.

NHF and hypercapnic respiratory failure

The physiological effects of NHF include improvement in carbon dioxide levels, as described earlier. When compared to NIV in COPD patients with moderate acute hypercapnic respiratory failure, there was no difference in treatment failure or mortality [174]. Home NHF in patients with chronic hypercapnic respiratory failure due to COPD reduced the rate of AECOPD and prolonged the time without AECOPD [175]. These studies suggest NHF may have clinical benefit in the treatment of patients with acute and/or chronic hypercapnic respiratory failure.

Summary

NIV, both bilevel and CPAP, has many indications across acute and chronic respiratory failure. When used appropriately, NIV has numerous physiological and clinical benefits, including improved gas exchange, improved respiratory muscle function, and improvements in breathlessness, quality of life and survival. In the future, technological advances in NIV devices will hopefully make them more user friendly, portable and customisable to meet specific individual patient needs. This should help to improve patient comfort and compliance with device use, thus leading to better patient outcomes. With improvements in the integration of data analytics with artificial intelligence, clinicians may be able to titrate NIV therapy more precisely to meet each patient's individual needs, thereby leading to optimisation of treatment outcomes and minimisation of complications. Moreover, advances in remote monitoring technologies will enhance the capabilities of telemonitoring and telemedicine programmes to provide real-time adjustments to treatment plans to address fluctuating changes in the patient's medical condition. This aspect is extremely important in dynamic diseases like COPD and will significantly increase the potential for patients to receive NIV therapy in the home to enhance their quality of life. The opportunities for an enhancement of NIV therapies to provide personalised medicine to meet the patient's needs for treatment of respiratory failure in a variety of settings outside of the hospital is bright, and hopefully will enhance the management of patients with advanced lung diseases who suffer from a range of disorders causing respiratory failure.

Shareable PDF

This one-page PDF can be shared freely online.

Shareable PDF ERJ-00396-2024.Shareable ^{(373.6KB, pdf)}