Page of

Non-invasive positive-pressure ventilation 

Non-invasive positive-pressure ventilation
Non-invasive positive-pressure ventilation
Oxford Textbook of Critical Care (2 ed.)

Giulia Spoletini

and Nicholas S. Hill


Key points

  • Exacerbations of chronic obstructive pulmonary disease (COPD), cardiogenic pulmonary oedema and acute respiratory failure in immunocompromised patients are the main indications for non-invasive ventilation (level 1 evidence).

  • Weaker evidence supports the use of non-invasive ventilation (NIV) in other forms of hypercapnic respiratory failure (acute asthma, cystic fibrosis, obesity–hypoventilation syndrome) and in hypoxaemic respiratory failure (pneumonia, acute respiratory distress syndrome (ARDS), chest trauma).

  • Proper selection of patients, considering both indications and contraindication to NIV, plays a key-role in NIV success.

  • Trained staff with knowledge and skill in the proper application of NIV is another determinant of success.

  • Avoid delays in the initiation of NIV as well as in intubation when NIV fails.


Non-invasive ventilation (NIV) is a technique that provides respiratory assistance and avoids airway invasion in patients with certain kinds of acute respiratory failure (ARF). The most common NIV application is non-invasive positive pressure ventilation (NIPPV), which conducts pressurized gas from a positive pressure ventilator to the patient via an external interface strapped over the nose and/or mouth. This technique avoids airway invasion, and can thereby avoid complications related to intubation and invasive mechanical ventilation (INV), especially ventilator-associated pneumonias. Furthermore, it avoids the trauma associated with intubation and usually augments comfort compared with INV, thereby reducing the need for analgesia and sedation. It is suitable for intermittent use, allowing breaks for eating and communicating. It may also reduce morbidity and mortality, and shorten ICU stays or avoid the need for ICU altogether, thus reducing costs. This brief review will focus on current indications for NIV, selection of appropriate patients, and proper application.

Indications with level 1 evidence: NIV is first choice for ventilatory support

Chronic obstructive pulmonary disease

The main indication for NIV in chronic obstructive pulmonary disease (COPD) is an exacerbation with persistent acute or acute on chronic hypercapnic respiratory failure (AHRF) despite initial medical treatment. In these patients, NIV combines positive end expiratory pressure (PEEP) to counterbalance auto-PEEP and pressure support to assist inspiratory muscles, thereby reducing the work of breathing and averting inspiratory muscle fatigue. Compared with standard medical care in COPD patients, NIV more rapidly improves vital signs (especially respiratory rate) and gas exchange, and reduces the need for intubation, complications, length of hospital stay, and mortality [1]‌. In particular, NIV is strongly recommended in those patients with a 7.25 < pH < 7.35. With a pH < 7.25, NIV could still be successful, but should be managed in an ICU with very close observation in order to intubate without delay if necessary [2,3].

Not only is NIV indicated for moderate to severe COPD exacerbations, but also for COPD patients with respiratory failure in a number of other settings including:

  • When associated with pneumonia.

  • Post-operative patients, especially after lung resection.

  • To facilitate extubation in COPD patients requiring INV initially.

  • To avoid extubation failure after a standard extubation.

  • In do-not-intubate patients.

Cardiogenic pulmonary oedema

In patients with cardiogenic pulmonary oedema (CPE), CPAP (10–12.5 cmH2O) compared with standard oxygen supplementation increases intrathoracic pressure and raises functional residual capacity, thus re-expanding alveoli, rapidly improving oxygenation, improving lung compliance, reducing left ventricular afterload, and often increasing cardiac output. These benefits translate into the avoidance of intubation and reduced mortality. In the past decade, a number of studies have compared ‘bilevel’ ventilation (combining inspiratory pressure support with positive expiratory pressure) with CPAP alone to treat CPE, but have found no differences between the modes in intubation, myocardial infarction and mortality rates [4]‌. The European Society of Cardiology considers CPAP as first-line treatment of CPE [5], but the Canadian NIV Guideline group considers them equivalent and recommends either CPAP or ‘bi-level’ ventilation for CPE patients in the absence of shock or need for coronary revascularization [6]. These positive results from in-patients have been replicated in the field where a number of randomized trials have demonstrated that CPAP administered by ambulance crews reduces intubation and mortality rates in CPE patients prior to hospitalization [7].

Immunocompromised patients

INV in immunocompromised patients is associated with infectious and haemorrhagic complications, including septic shock and high mortality rates. NIV can reduce the occurrence of nosocomial infections and septic shock and avoid INV in such patients, thereby improving survival [8]‌. It should be emphasized, however, that such patients should be started early on NIV, closely monitored in an ICU, and promptly intubated if they continue to deteriorate with worsening oxygenation and haemodynamic instability.

Weaker evidence—NIV is an option

Other forms of hypercapnic respiratory failure


Some earlier uncontrolled cohort series have associated improvements in gas exchange and low intubation rates with use of NIV in patients with severe asthma, but no controlled trials have confirmed these benefits. Some of the few randomized trials reported have demonstrated that NIV increases airflow more rapidly than a sham mask [9]‌ or standard oxygen [10] during severe asthma attacks, suggesting a bronchodilator effect of positive pressure. One study also showed a lower hospitalization rate from the ED with NIV [9]. Thus, NIV with close monitoring is an option for severe asthma that does not respond promptly to standard medical treatment, perhaps combined with helium oxygen mixtures and/or continuous bronchodilator nebulization. However, considering that supportive evidence is weak, patients should be monitored closely and promptly intubated if they fail to improve.

Cystic fibrosis

NIV has been used for cystic fibrosis in the chronic or subacute setting as a bridge-to-transplant [11]. It may also serve a role as an adjuvant to secretion clearance techniques. However, few studies have evaluated its use in acute care. It remains an option in the acute setting, but for patients with acute pneumonias and secretion retention, prompt intubation followed by transition to NIV once the patient has stabilized may be the preferred approach.

Obesity-hypoventilation syndrome

Morbid obesity is often associated with chronic hypoventilation, defining the obesity-hypoventilation syndrome (OHS). Episodes of acute on chronic respiratory acidosis occurring in this entity may be treated with NIV, avoiding INV. Some of these patients are initially too ill for NIV, due to altered neurological status, excessive secretions, or decompensated congestive heart failure, but can be transitioned to NIV when stabilized. In one cohort, initiation during acute exacerbations, rather than the chronic state, and cardiovascular morbidities were associated with worse long-term NIV outcomes [12].

Hypoxaemic respiratory failure


Pneumonia and ARDS are associated with a substantially higher risk of NIV failure than most other forms of acute respiratory failure, and NIV is not routinely recommended to treat them. Oxygenation needs are high, necessitating higher PEEP levels, which detracts from mask comfort and predisposes to greater mask leak. In addition, the underlying process often progresses, rendering the patient too ill to be safely managed with NIV. Thus, patients with ARDS or pneumonia treated with NIV must be selected with great caution. They must not have multi-organ dysfunction or haemodynamic instability, and must be watched very carefully in an ICU. If their oxygenation does not improve substantially within 1–2 hours of NIV initiation, they should be promptly intubated [13].

Chest trauma

Chest trauma patients may develop hypoxaemic ARF after lung contusion or acute lung injury related to activation of the inflammatory cascade. They may also develop hypercapnic ARF after rib fractures, leading to flail chest. A randomized controlled trial showed avoidance of respiratory muscle fatigue and the need for intubation in severely hypoxemic chest trauma patients [14]. A subsequent meta-analysis [15] showed that NIV not only reduces intubation rate, but also length of ICU stay and mortality, and may be a preferable alternative to INV in these patients.


Many patients with chronic respiratory conditions decide to forego heroic measures and decline intubation. NIV can still help many of these patients, especially those with COPD or CPE who have a reasonably good chance of surviving hospitalization for an acute exacerbation [16]. NIV may also be used for palliative purposes, mainly to relieve dyspnoea in terminal patients, or to extend life long enough for relatives to visit or settle their affairs. However, it is important to make certain that patients and their families who wish for NIV understand such circumstances and establish realistic goals for its use. If these goals are not being met, NIV should be stopped and alternative palliation instituted.

Practical aspects

The success of NIPPV depends not only on the disease underlying NIV, but also on the proper selection of patients, timing, and setting of the treatment.

Selection of patients

All patients in respiratory distress, defined as tachypnoea, use of accessory muscles, hypoxaemia, diagnosed with ARF (pH < 7.35 and pCO2 > 45 mmHg, or PO2/FiO2 < 200), should be considered as soon as possible for NIV. Delay in starting the treatment can increase the risk of complications and NIV failure. Before starting NIV, it is mandatory to determine whether they have contraindications, such as respiratory arrest, septic shock with multiple organ failure, or the inability to fit the mask or to protect the airways, in which case they should be promptly intubated. Some patients have relative contraindications, such as unstable medical conditions, agitation, the presence of excessive secretions, or recent thoracic or upper abdomen surgery, and should be evaluated on a case-by-case basis. In marginal cases at higher risk of NIV failure, a NIV trial of 1–2 hours could be tried and intubation undertaken if there is no improvement [17].


The choice of a comfortable and properly fitted interface is key to a favourable NIV outcome, since it can improve patients’ tolerance and adherence to the treatment, minimize air leaks, and optimize patient–ventilator synchrony. Oro-nasal (or full-face) are generally preferred, since they are associated with less air leaking through the mouth. On the other hand, nasal masks can facilitate speech and expectoration, and may be better tolerated by claustrophobic patients. The helmet is an alternative interface that consists of a soft clear plastic cylinder that seals over the neck and shoulders, and has been extensively studied, mainly in Italy. It has proven quite effective in delivering CPAP, but requires high air flow to minimize rebreathing, which are quite noisy. When used to deliver pressure support and PEEP, patient–ventilator asynchrony can be a problem with the helmet [18], but a newer, stiffer version may counter this problem.

Ventilator selection

Either ICU ventilators using pressure- or volume-limited modes or portable bi-level ventilators can be used for NIV with equal expectations of success. However ‘bi-level’ devices designed specifically for delivery of NIV are the most often chosen for hypercapnic ARF, whereas ICU ventilators may be selected more often for hypoxaemic respiratory failure. Many ICU ventilators now have ‘NIV’ modes that facilitate delivery of NIV, but may require additional adjustments in the face of leaks.

Ventilator settings

Successful NIV requires that sufficient ventilator support is provided in synchrony with the patient’s breathing efforts. Thus, proper adjustment of ventilator settings is very important. Usually, the pressure-support (or bi-level) mode starting with lower settings (inspiratory pressure 8–12 cmH2O, expiratory pressure 4–5 cmH2O) enhances tolerance for patients. Inspiratory pressure can then be gradually adjusted upward to alleviate persistent respiratory distress, within the patient’s limit of tolerance. Expiratory pressure can be increased to improve triggering in the presence of auto-PEEP or to improve oxygenation. Oxygen supplementation is titrated to achieve a desired O2 saturation (>90–92%). Humidification is not necessary for brief applications (such as for CPE), but helps to avoid mucosal drying and enhances comfort for longer applications [18].


Close monitoring of NIV is important for success, especially during the early adaptation period. The value of frequent checks to optimize mask fit and ventilator settings, and to encourage the patient cannot be overemphasized. Acutely-ill patients should be monitored in an ICU or intermediate care unit until they are more stable. NIV can be administered on regular wards by experienced staff, but only for cooperative patients who are relatively stable and can call for help if needed [19].

All patients undergoing NIV should be monitored at least with:

  • Clinical examination of:

    • Dyspnoea.

    • Respiratory and heart rate.

    • Other signs of respiratory distress—abdominal paradox, accessory muscles use, etc.).

    • Mask comfort, fit.

  • Gas exchange:

    • Arterial blood gases before initiation, after 1–2 hours of NIV and as needed.

    • Continuous O2 saturation.

  • Blood pressure and EKG.

  • Ventilator:

    • Synchrony with patient.

    • Tidal volume (target 6–7 mL/kg ideal body weight).

    • Air leaks from mask and tubing.


NIPPV is usually a safe and well-tolerated technique. The adverse effect are related to the interface, which can cause skin reddening or ulceration, especially over the nasal bridge, claustrophobia, sinus or nasal pain, and dryness of eyes and mouth due to air leaks. Improbable, but possible complications include pneumothoraces, painful gastric insufflation, and aspiration of gastric contents. Patient agitation interferes with efficacy, but can be ameliorated by sedation.

Predicting outcome

When applied in properly selected patients using optimal techniques by an experienced staff, success rates of NIV, especially for COPD and CPE, are generally over 80%. However, some patients fail, especially if they have diagnoses like pneumonia or cancer. In hypercapnic ARF, APACHE II score > 34, pH < 7.25, respiratory rate > 35, and Glasgow Coma Score ≤ 11 have all been shown to predict NIV failure, especially if they persist for the first 2 hours [17]. In hypoxaemic ARF, having ARDS, pneumonia, or shock are the strongest predictors of failure, but other predictors include older age, high acuity of illness, and lack of improvement in PaO2/FiO2 after the first hour of NIV [20].


NIV has been one of the most important developments in the field of mechanical ventilation over the past 20 years and its use is increasing around the world. This increase has been associated with improved outcomes for patients with ARF due to COPD and CPE, and it is being used increasingly in other types of ARF as well. To achieve optimal results, it must be applied in carefully-selected patients, using appropriate techniques implemented by a skilled staff with conscientious monitoring.


1. Quon BS, Gan WQ, and Sin DD. (2008). Contemporary management of acute exacerbations of COPD: a systematic review and meta-analysis. Chest, 133(3), 756–66.Find this resource:

2. Plant PK, Owen JL, and Elliot MW. (2000). Early use of non-invasive ventilation for acute exacerbations of chronic obstructive pulmonary disease on general respiratory ward: a multicentre randomized trial. Lancet, 355, 1931–5.Find this resource:

3. Squadrone E, Frigerio P, Fogliati C, et al. (2004). Noninvasive vs invasive ventilation in COPD patients with severe acute respiratory failure deemed to require ventilatory assistance. Intensive Care Medicine, 30, 1303–10.Find this resource:

4. Masip J, Roque M, Sanchez B, et al. (2005). Noninvasive ventilation in acute cardiogenic pulmonary edema: systematic review and meta-analysis. Journal of the American Medical Association, 294(24), 3124–30.Find this resource:

5. The Task Force on Acute Heart Failure of the European Society of Cardiology. (2005). Executive summary of the guidelines on diagnosis and treatment of acute heart failure. European Heart Journal, 26, 384–416.Find this resource:

6. Keenan SP, Sinuff T, Burns KE, et al. (2011). Canadian Critical Care Trials Group. Clinical practice guidelines for the use of noninvasive positive-pressure ventilation and noninvasive continuous positive airway pressure in the acute care setting. Canadian Medical Association Journal, 183(3), E195–214.Find this resource:

7. Ducros L, Logeart D, Vicaut E, et al. (2011). CPAP for acute cardiogenic pulmonary oedema from out-of-hospital to cardiac intensive care unit: a randomized multicenter study. Intensive Care Medicine, 37(9), 1501–9.Find this resource:

8. Hilbert G, Gruson D, Vargas F, et al. (2001). Noninvasive ventilation in immunosuppressed patients with pulmonary infiltrates, fever, and acute respiratory failure. New England Journal of Medicine, 344, 481–7.Find this resource:

9. Soroksky A, Stav D, and Shpirer I. (2003). A pilot prospective, randomized, placebo-controlled trial of bilevel positive airway pressure in acute asthmatic attack. Chest, 123, 1018–25.Find this resource:

10. Soma T, Hino M, Kida K, and Kudoh S. (2008). A prospective and randomized study for improvement of acute asthma by non-invasive positive pressure ventilation (NPPV). Internal Medicine, 47(6), 493–501.Find this resource:

11. Noone PG. (2008). Non-invasive ventilation for the treatment of hypercapnic respiratory failure in cystic fibrosis. Thorax, 63, 5–7.Find this resource:

12. Borel JC, Burel B, Tamisier R, et al. (2013). Comorbidities and mortality in hypercapnic obese under domiciliary noninvasive ventilation. PloS One, 8, e52006.Find this resource:

13. Antonelli M, Conti G, Esquinas A, et al. (2007). A multiple-center survey on the use in clinical practice of noninvasive ventilation as a first-line intervention for acute respiratory distress syndrome. Critical Care Medicine, 35(1), 18–25.Find this resource:

14. Hernandez G, Fernandez R, Lopez-Reina P, et al. (2010). Noninvasive ventilation reduces intubation in chest trauma-related hypoxemia. Chest, 137, 74–80.Find this resource:

15. Chiumello D, Coppola S, Froio S, Gregoretti C, and Consonni D.(2013). Noninvasive ventilation in chest trauma: a systematic review and meta-analysis. Intensive Care Medicine, 39, 1171–80.Find this resource:

16. Levy M, Tanios MA, Nelson D, et al. (2004). Outcomes of patients with do-not-intubate orders treated with noninvasive ventilation. Critical Care Medicine, 32, 2002–7.Find this resource:

17. Confalonieri M, Garuti G, Cattaruzza MS, et al. (2005). A chart of failure risk for noninvasive ventilation in patients with COPD exacerbation. European Respiratory Journal, 25, 348–55.Find this resource:

18. Nava S, Navalesi P, and Gregoretti C. (2009). Interfaces and humidification for noninvasive mechanical ventilation. Respiratory Care, 54(1), 71–84.Find this resource:

19. Elliott MW, Confalonieri M, and Nava S. (2002). Where to perform noninvasive ventilation? European Respiratory Journal, 19, 1159–66.Find this resource:

20. Antonelli M, Conti G, Moro ML, et al. (2001). Predictors of failure of noninvasive positive pressure ventilation in patients with acute hypoxemic respiratory failure: a multi-center study. Intensive Care Medicine, 27, 1718–28.Find this resource:

Copyright © 2022. All rights reserved.