ERS Practical Handbook of Noninvasive Ventilation 9781849840767, 1849840768, 9781849840750

Citation preview

practical Handbook

Noninvasive Ventilation Editor Anita K. Simonds

PUBLISHED BY THE EUROPEAN RESPIRATORY SOCIETY

CHIEF EDITOR Anita K. Simonds (London, UK)

ERS STAFF Eddy Baker, Alice Bartlett, Matt Broadhead, Lee Dodd, Jonathan Hansen, Catherine Pumphrey, Elin Reeves, David Sadler, Claire Turner, Ben Watson © 2015 European Respiratory Society Cover image: Juanmonino, iStockphoto Design by Claire Turner, Lee Dodd, ERS Typeset by Techset Composition Ltd Printed by Latimer Trend Indexed by Merrall-Ross International Limited All material is copyright to the European Respiratory Society. It may not be reproduced in any way including electronically without the express permission of the society. CONTACT, PERMISSIONS AND SALES REQUESTS: European Respiratory Society, 442 Glossop Road, Sheffield, S10 2PX, UK Tel: +44 114 2672860 Fax: +44 114 2665064 e-mail: [email protected]

ISBN 978-1-84984-075-0

Table of contents Contributors Preface Get more from this Practical Handbook List of abbreviations

vii xi xii xiii

Chapter 1 – Introduction NIV: past, present and future Anita K. Simonds

1

Chapter 2 – Getting the basics right: equipment Basic principles of ventilators Alanna Hare and Michelle Chatwin

10

Matching mode and settings to the patient: an introduction Jan H. Storre and Jens C. Callegari

18

Choosing the interface Anne-Kathrin Brill

26

Supplemental oxygen and humidication Sundeep Kaul and Anita K. Simonds

35

Chapter 3 – Getting the basics right: patient selection The patient with an acute hypercapnic exacerbation of COPD Mark W. Elliott

41

Patients with acute hypercapnic respiratory failure and neuromuscular or chest wall disease Anita K. Simonds

49

Patients with acute hypercapnic respiratory failure and OHS Nicholas Hart and Patrick B. Murphy

56

Patients with acute-on-chronic hypercapnic respiratory failure 60 due to non-COPD obstructive lung disease and interstitial disorders Marieke L. Duiverman and Peter J. Wijkstra The patient with acute hypoxaemic respiratory failure excluding pulmonary oedema Pongdhep Theerawit, Yuda Sutherasan and Paolo Pelosi

67

The patient with acute hypoxaemic failure and cardiogenic pulmonary oedema João C. Winck and Luís F. Azevedo

72

Chapter 4 – Paediatric indications for NIV Acute NIV in children, including ventilator and interface choice Alessandro Amaddeo and Brigitte Fauroux

79

Chapter 5 – Airway clearance and physiotherapy Airway clearance methods and nebulised therapy in acute NIV Michelle Chatwin

86

Chapter 6 – Monitoring progress in acute NIV Monitoring choices in acute NIV Raffaele Scala

93

Starting and stopping acute NIV: when and why? Bernd Schönhofer

102

Problem-solving: case studies of NIV problems and their

111

management

Alanna Hare

Chapter 7 – NIV and the intensive care unit NIV and weaning Miquel Ferrer

118

NIV to avoid re-intubation Paolo Navalesi and Federico Longhini

127

NIV in the perioperative period Yuda Sutherasan, Maria Vargas and Paolo Pelosi

135

NIV for endoscopic procedures Leo M.A. Heunks, Lisanne Roesthuis and Erik H.F.M. van der Heijden

142

NIV in respiratory pandemics Anita K. Simonds

148

Stepping up and down from NIV to tracheostomy ventilation Mark W. Elliott

155

Chapter 8 – Long-term NIV Chronic NIV in hereditary neuromuscular disorders Anita K. Simonds

163

Chronic NIV in motor neurone disease/ALS Joan Escarrabill

176

Chronic NIV in chest wall disorders Marieke L. Duiverman and Peter J. Wijkstra

182

Chronic NIV in COPD Wolfram Windisch and Jan H. Storre

190

Chronic NIV in OHS Patrick B. Murphy and Nicholas Hart

197

Chronic NIV in bronchiectasis, CF and interstitial lung disease Amanda J. Piper

204

Chronic NIV in heart failure patients: ASV, NIV and CPAP João C. Winck, Marta Drummond, Miguel Gonçalves and Tiago Pinto

211

Chronic NIV in children: indications, outcomes and transition Hui-Leng Tan and Anita K. Simonds

217

Practicalities of and guide to cough augmentation and daytime mouthpiece ventilation Michelle Chatwin

226

Long-term NIV failure: causes and problem solving Jean-Paul Janssens, Dan Adler and Jesus Gonzalez-Bermejo

234

Long-term NIV case histories Alanna Hare

246

Chapter 9 – NIV for symptom palliation NIV in palliative care and at the end of life Anna Maria Cuomo

253

Chapter 10 – Discharge planning and community care Discharging the ventilator-dependent adult and child Joan Escarrabill

260

Home monitoring and follow-up of long-term NIV Dan Adler, Claudio Rabec and Jean-Paul Janssens

265

Assessing quality of life and outcome of long-term NIV Sophie E. Huttmann and Wolfram Windisch

276

Patient and carer education, and risk management Joan Escarrabill

282

Chapter 11 – Setting up an NIV service Setting up and staffing your NIV unit Joan Escarrabill

289

Index

295

Contributors Chief Editor Anita K. Simonds NIHR Respiratory Biomedical Research Unit, Royal Brompton and Hareeld NHS Foundation Trust, London, UK [email protected] Authors Dan Adler Division of Pulmonary Diseases, Geneva University Hospital, Geneva, Switzerland [email protected] Alessandro Amaddeo Pediatric Noninvasive Ventilation and Sleep Unit, Necker University Hospital, Paris, France [email protected] Luís F. Azevedo Dept of Health Information and Decision Sciences, Center for Health Technology and Services Research – CINTESIS (Centro de Investigação em Tecnologias e Serviços de Saúde), Faculdade de Medicina da Universidade do Porto, Portugal [email protected] Anne-Kathrin Brill University Hospital Berne, Inselspital, Dept of Pneumology, Bern, Switzerland [email protected]

Jens C. Callegari Cologne Merheim Hospital, Dept of Pneumology, Kliniken der Stadt Köln gGmbH, Witten/Herdecke University, Cologne, Germany [email protected] Michelle Chatwin Clinical and Academic Department of Sleep and Breathing, Royal Brompton and Hareeld NHS Foundation Trust, London, UK [email protected] Anna Maria Cuomo Freelance oncologist, Bologna, Italy [email protected] Marta Drummond Pulmonology Dept, Faculty of Medicine, Porto University, Porto, and Sleep and Non-invasive Ventilation Unit, Pulmonology Dept, Centro Hospitalar São João, Porto, Portugal [email protected] Marieke L. Duiverman Department of Pulmonology/Home Mechanical Ventilation, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands [email protected] Mark W. Elliott Dept of Respiratory Medicine, Sleep and Non-invasive Ventilation Services, St. James’s University Hospital, Leeds, UK [email protected]

Joan Escarrabill Hospital Clínic, Chronic Care Program, Barcelona, Spain [email protected] Brigitte Fauroux Pediatric Noninvasive Ventilation and Sleep Unit, Necker University Hospital, Paris, France [email protected] Miquel Ferrer Respiratory Intensive and Intermediate Care Unit, Dept of Pneumology, Thorax Institute, Hospital Clínic, IDIBAPS, Barcelona, Spain [email protected] Miguel Gonçalves Pulmonology Dept, Faculty of Medicine, Porto University, and Sleep and Non-invasive Ventilation Unit, Pulmonology Dept, Centro Hospitalar São João, Porto, Portugal [email protected] Jesus Gonzalez-Bermejo Service de Pneumologie et Réanimation, GH Pitié Salpêtrière, Paris, France [email protected] Alanna Hare Royal Brompton and Hareeld NHS Foundation Trust, London, UK [email protected]

Nicholas Hart Lane Fox Clinical Respiratory Physiology Research Centre, Guy’s and St Thomas’ NHS Foundation Trust, Division of Asthma, Allergy and Lung Biology, King’s College London, and Lane Fox Respiratory Unit, Guy’s and St Thomas’ NHS Foundation Trust, London, UK [email protected] Leo M.A. Heunks Dept of Intensive Care Medicine, Radboud University Medical Centre Nijmegen, Nijmegen, The Netherlands [email protected] Sophie E. Huttmann Cologne Merheim Hospital, Department of Pneumology, Kliniken der Stadt Köln gGmbH Witten/Herdecke University, Faculty of Health/School of Medicine, Cologne, Germany [email protected] Jean-Paul Janssens Division of Pulmonary Diseases, Geneva University Hospital, Geneva, Switzerland [email protected] Sundeep Kaul Royal Brompton and Hareeld NHS Foundation Trust, Hareeld Hospital, Hareeld, UK [email protected] Federico Longhini Anesthesia and Intensive Care, Sant’Andrea Hospital, ASL VC, Vercelli, Italy [email protected]

Patrick B. Murphy Lane Fox Clinical Respiratory Physiology Research Centre, Guy’s and St Thomas’ NHS Foundation Trust, Division of Asthma, Allergy and Lung Biology, King’s College London, and Lane Fox Respiratory Unit, Guy’s and St Thomas’ NHS Foundation Trust, London, UK [email protected] Paolo Navalesi Anesthesia and Intensive Care, Sant’Andrea Hospital, ASL VC, Vercelli, Dept of Translational Medicine, Eastern Piedmont University “A. Avogadro”, Novara, and CRRF Mons. L. Novarese, Moncrivello, Italy [email protected] Paolo Pelosi IRCCS AOU San Martino-IST, Dept of Surgical Sciences and Integrated Diagnostics, University of Genoa, Genoa, Italy [email protected] Tiago Pinto Sleep and Non-invasive Ventilation Unit, Pulmonology Dept, Centro Hospitalar São João, Porto, Portugal [email protected] Amanda J. Piper Dept of Respiratory and Sleep Medicine, Royal Prince Alfred Hospital, Camperdown, and Woolcock Institute of Medical Research, University of Sydney, Sydney, Australia [email protected]

Claudio Rabec Service de Pneumologie et Réanimation, Centre Hospitalier et Universitaire de Dijon, Dijon, France [email protected] Lisanne Roesthuis Dept of Intensive Care Medicine, Radboud University Medical Centre Nijmegen, Nijmegen, The Netherlands [email protected] Raffaele Scala Pulmonary Unit and Respiratory Intensive Care Unit, S. Donato Hospital, Arezzo, Italy [email protected] Bernd Schönhofer Pneumologie, Internistische Intensivmedizin und Schlafmedizin, KRH Klinikum Siloah- OststadtHeidehaus, Hannover, Germany [email protected] Jan H. Storre Cologne Merheim Hospital, Dept of Pneumology, Kliniken der Stadt Köln gGmbH, Witten/Herdecke University, Cologne, and Dept of Pneumology, University Hospital Freiburg, Germany [email protected] Yuda Sutherasan Ramathibodi Hospital, Mahidol University, Bangkok, Thailand [email protected] Hui-Leng Tan Dept of Paediatric Respiratory Medicine, Royal Brompton Hospital, Royal Brompton and Hareeld NHS Foundation Trust, London, UK [email protected]

Pongdhep Theerawit Ramathibodi Hospital, Mahidol University, Bangkok, Thailand [email protected] Erik H.F.M. van der Heijden Dept of Pulmonary Diseases, Radboud University Medical Centre Nijmegen, Nijmegen, The Netherlands [email protected] Maria Vargas Dept of Neuroscience and Reproductive and Odontostomatological Sciences, University of Naples “Federico II”, Naples, Italy [email protected] Peter J. Wijkstra Department of Pulmonology/Home Mechanical Ventilation, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands [email protected]

João C. Winck Pulmonology Dept, Center for Health Technology and Services Research – CINTESIS (Centro de Investigação em Tecnologias e Serviços de Saúde), Faculdade de Medicina da Universidade do Porto, Porto, Portugal [email protected] Wolfram Windisch Cologne Merheim Hospital, Dept of Pneumology, Kliniken der Stadt Köln gGmbH Witten/Herdecke University, Faculty of Health/School of Medicine, Cologne, Germany [email protected]

Preface Bearing in mind the growing applications for noninvasive ventilation (NIV) and the success of the ERS NIV courses, we felt the time had come for an ERS Practical Handbook of Noninvasive Ventilation. Gaining knowledge in NIV is not difficult as there are many resources available – what makes the difference in clinical practice is expertise in applying the technique and acquiring skills in problem-solving. It is these practical issues this book attempts to address, together with providing an up-to-date guide to which patients to select for therapy, and the evidence on which this is based. NIV is a rapidly changing eld. In recent months we have had trial results on the use of adaptive servo ventilation, suggesting that this does not improve outcome in heart failure patients with central sleep apnoea; and indications for home NIV in chronic COPD have been informed by recent trial results. One of the greatest values of ERS NIV courses is the chance to discuss challenges in patient care. Some of these are perennial – persuading a confused, hypercapnic patient or small child to cope with a mask, avoiding interface leaks, and balancing ventilatory needs with patient tolerance. In this book we have invited those in the front line of NIV to share their views and experiences. This means there will be a variety of approaches suggested in some cases, whereas in others there may be a greater degree of consensus, or even complete agreement. We very much hope this will help the reader not only to make the right decision for their patients but also to gain condence from the very practical advice offered. I am extremely grateful to all the contributors and reviewers, and to the ERS Publications Office team for their great support and enthusiasm.

Anita K. Simonds Chief Editor

Get more from this Handbook By buying the ERS Practical Handbook of Noninvasive Ventilation, you also gain access to the electronic version of the book, as well as an accredited online CME test.

Simply visit http://erspublications.com/titles and add the ERS Practical Handbook of Noninvasive Ventilation to your cart. At the checkout, enter the unique voucher code printed on the inside front cover of the book. You will then be able to download the entire book in PDF format to read on your computer or mobile device. You’ll also be able to take the online CME test. This Practical Handbook has been accredited by the European Board for Accreditation in Pneumology (EBAP) for 12 CME credits. Also available from the ERS ERS Handbook: Self-Assessment in Respiratory Medicine Edited by Konrad E. Bloch, Anita K. Simonds and Thomas Brack The new and updated second edition of Self-Assessment in Respiratory Medicine is an invaluable tool for any practitioner of adult respiratory medicine. The 261 multiple-choice questions cover the full breadth of the specialty, using clinical vignettes that test not only readers’ knowledge but their ability to apply it in daily practice.

List of abbreviations AHI AIDS ALS ARDS ARF ASB ASV ASSPCV AVAPS BMI CF COPD CPAP ECG EPAP FEV1 FIO2 FVC HIV ICU IPAP IPPV IVAPS NIV NPV OHS OSA(S) PaCO2 PaO2 PAV PCV PEEP PSV PtcCO2 SaO2 RCT SpO2 TB TLC VCV VT

Apnoea–hypopnoea index Acquired immunodeciency syndrome Amyotrophic lateral sclerosis Acute respiratory distress syndrome Acute respiratory failure Assisted spontaneous breathing Adaptive servo ventilation Assisted pressure-controlled ventilation Average volume-assured pressure support Body mass index Cystic brosis Chronic obstructive pulmonary disease Continuous positive airway pressure Electrocardiogram Expiratory positive airway pressure Forced expiratory volume in 1 s Inspiratory oxygen fraction Forced vital capacity Human immunodeciency virus Intensive care unit Inspiratory positive airway pressure Intermittent positive pressure ventilation Intelligent volume-assured pressure support Noninvasive ventilation Negative pressure ventilation Obesity hypoventilation syndrome Obstructive sleep apnoea (syndrome) Arterial carbon dioxide tension Arterial oxygen tension Proportional assist ventilation Pressure-controlled ventilation Positive end-expiratory pressure Pressure support ventilation Transcutaneous carbon dioxide tension Arterial oxygen saturation Randomised controlled trial Arterial oxygen saturation measured by pulse oximetry Tuberculosis Total lung capacity Volume-controlled ventilation Tidal volume

Introduction

NIV: past, present and future

Anita K. Simonds The history of NIV is an intertwining chronicle of the development of negative and positive pressure modes, as at different times in history, each mode has dominated. For example, in the mid-1980s, all patients using respiratory support received NPV, whereas the vast majority of our patients now use positive-pressure NIV, and there are hundreds of thousands of individuals with sleep apnoea worldwide receiving CPAP and many thousands using NIV for respiratory failure. Indeed, the history of NIV dates right back to the beginning: “And the LORD God formed man of the dust of the ground, and breathed into his nostrils the breath of life; and man became a living soul”. Genesis 2: 7 So NIV existed before invasive ventilation! More scientifically, from an invasive ventilation perspective, there are descriptions going back to antiquity of tracheostomy, and these are found in the medieval period and 1500s too. Vesalius, in the 16th century, was aware that positive pressure applied to the trachea would inflate the lungs and Hooke demonstrated that it was possible to keep a dog alive by applying bellows to the upper airway in 1667. For verifiable human accounts, we need to advance to the 18th century – and first deal with NPV. Woollam (1976) credits John Dalziel, a Scot, as the first to describe a tank ventilator/iron lung type device and a fellow Scot, Alexander Graham Bell, developed the notion further. Bell is best known for inventing the forerunner of the telephone in 1876 but, in 1881, his first child Edward died of respiratory distress a few hours after birth. The story goes that shortly afterwards Bell was walking along a shingle beach in Ontario when the idea of a negative-pressure jacket to assist the breathing of infants came to him. He went on to patent this device. Figure 1 is an extract from his notebook, although it is quite difficult to decipher.

Key points • The development of negative- and positive-pressure NIV is inextricably linked. • NIV is one of the most evidence-based areas of respiratory medicine and indications for NIV continue to increase in number. ERS Practical Handbook Noninvasive Ventilation

1

Introduction

Figure 1.  Drawing of a negative pressure device for children by Alexander Graham Bell. Image: US Library of Congress, Washington, DC, USA.

In fairness, similar ideas were flourishing elsewhere, in Europe, but it was not until the 1920s that an iron lung with a motorised pump was developed by Drinker in 1928 in the USA and demonstrated in London in 1931. This concept then entered the medical mainstream, disseminated by a brisk correspondence in the Lancet. That was fortuitous, as the coming scourge was epidemics of poliomyelitis, which had begun in the First World War and swept across Europe and the USA in the 1930s–1950s. Polio paralyses the respiratory muscles as well as limbs muscles, resulting in respiratory failure. Iron lungs were pressed into action, including the intimidating multitier versions in figure 2. There is no doubt that iron lungs saved thousands of lives but they were big, cumbersome and expensive (the original Drinker ventilator cost $1500 – equivalent to the cost of a US new-build house at the time), and so were not going to be a practical way forward in respiratory care. In their observations on the use of negative-pressure respirators in polio, Plum and Wolff (1951) found that the tank ventilator was safest for managing respiratory insufficiency and that in the acute phase of polio, the cuirass was too inefficient. Upper airway obstruction provoked by the negative pressure was a common problem. Practical limitations were compounded by the huge outbreak of polio in Denmark in 1952, which was associated with a very high prevalence of cases with bulbar weakness. Not only was an insufficient number of iron lungs available but these were also inadequate in caring for patients with bulbar problems – mortality rose to 90% and the only solution open to Ibsen (1954) and the Danish anaesthetic and medical teams was invasive positive-pressure ventilation via a tracheostomy

2

ERS Practical Handbook Noninvasive Ventilation

Introduction

Figure 2.  Multitier iron lung used in poliomyelitis epidemics. Reproduced from Kacmarek (2011) Respir Care; 56: 1170–1180 with permission from the publisher.

or endotracheal tube. This switch to positive pressure continued and iron lungs began to disappear, heralding the arrival of the modern ICU. There was a brief resurgence of NPV in the 1970s and 1980s, but mainly to care for those with chronic ventilatory failure. Turning to the development of noninvasive positive pressure, this started at a slightly earlier time. Possibly the first well documented use of mask ventilation, in the 1760s, was in “resuscitation boxes”, which contained bellows to insufflate the lungs, tubing and glass nasal masks, and were placed by the Royal Humane Society (London, UK) to be used in the rescue of “drowned persons”. The first was located by the Serpentine Lake in Hyde Park, London. Ice skating was much in vogue at that time and the Serpentine froze in the winter. It seems children frequently fell through the ice and had to be rescued; there is even a protocol for the resuscitation of children rescued from a frozen lake. In retrospect, it is difficult to think of a better prognostic group to resuscitate – young, fit and cooled, providing they were retrieved quickly enough. But this was all manually applied positive-pressure ventilation and the first true motorised ventilator did not appear until the turn of the 20th century. In May 1908, under the headline “Smother small dog to see it revived”, The New York Times described a demonstration to the King County Medical Society, in Brooklyn, NY, USA, of a mechanised ventilator developed by Prof. George Poe (spookily related to mystery writer Edgar Allan Poe). A young boy was given a quarter to find a stray dog on the streets and this “cur” was smothered till apparently lifeless and then successfully resuscitated with the ventilator, to the acclaim of the audience! So, a good day for ventilators, but a bad day for stray dogs in Brooklyn. The best description of use in real clinical practice comes from Germany with the Dräger Pulmotor (Drägerwerk AG, Lubeck, Germany). This was patented by Heinrich Dräger in 1907 and was an innovative time-cycled device that delivered positive pressure during inspiration and negative pressure during expiration. However, it had a flaw in that the mask was connected to the ventilator by a ERS Practical Handbook Noninvasive Ventilation

3

Introduction

single limb of tubing, meaning that the carbon dioxide in the exhaled breath was rebreathed, which could eventually result in asphyxiation. Fortunately, Heinrich’s son Bernard redesigned the circuitry with two sets of tubing, one for inhalation, the other for exhalation, which solved that problem; this modification went into production and 30 years later, the 12 000th Pulmotor rolled off the production line in Lubeck. The Pulmotors were supplied to mines for poisoning accidents, to deal with victims of fires and for other acute uses. This is crucial, as the stimulus for ventilator use had been entirely for ARF up to this point. It was not until the 1970s and 1980s that long-term chronic use began to be the spur to ventilatory progress. This was partly related to better understanding of the physiology of breathing during sleep, the rediscovery of sleep apnoea and CPAP therapy, and underlying global trends in the switch from acute to chronic healthcare. OSA is associated with recurrent episodes of upper airway obstruction, which can lead to a number of vascular complications if not addressed, but which Sullivan showed in 1981 could be effectively treated with CPAP, as the airflow splints the airway open. In addition, developments in masks and technology extended NIV to respiratory failure in patients with neuromuscular disease. The original CPAP machines were very large – about the size of a vacuum cleaner – but have improved, and become smaller and portable over time. Importantly, mask design and comfort have improved too. To complete the timeline, the developments from the end of the 1980s to the present are shown in figure 3. A great deal of progress has occurred such that NIV is now one of the most evidence-based areas of respiratory medicine, as this handbook will describe. Really significant interventions are the discovery and confirmation by RCT that NIV halves mortality and morbidity in acute exacerbations of COPD, and this

CPAP, neuromuscular disease

NIV in restrictive disorders

1980s

Chronic

Improved outcome in Duchenne MD

1990s

NIV in motor neurone NIV in disease chronic COPD Paediatric NIV

Late 1990s

2000

2005

2014

NIV in acute hypoxaemic respiratory failure, pulmonary oedema and weaning

NIV in acute COPD

Acute

Figure 3.  Timeline of developments in NIV from the 1980s to the present day. MD: muscular dystrophy.

4

ERS Practical Handbook Noninvasive Ventilation

Introduction

provides the rationale for NIV to be available in every acute unit that admits respiratory patients, and for NIV to be used post-operatively in high-risk patients and for weaning. An additional major change in the past 30–40 years has been the increasing indications for long-term, chronic NIV and, of course, long-term application of CPAP in OSA. For patients with a range of causes of ventilatory failure, the natural history progresses from normal breathing, to a gradual loss in lung volumes and then, initially, changes in blood gases are seen at night due to hypoventilation, and if that is not addressed, ultimately, progression to daytime respiratory failure, cardiac decompensation and premature death. The interval between the onset of respiratory failure and death may be as short as a few years. In Duchenne muscular dystrophy, once a patient has developed a raised carbon dioxide level during the day, there is a 90% chance that they will be dead within a year. Figure 4 shows the long-term outcome of different groups of patients treated with NIV having developed severe ventilatory failure or progressed to cor pulmonale pretreatment. In post-polio patients, 5-year survival with NIV is 100% and it appears that these individuals will live to their normal life expectancy. 5-year survival is ∼80% in the other restrictive conditions. Results are less good in COPD and bronchiectasis for two reasons: these are intrinsic lung disease conditions rather than being restrictive disorders with normal lungs, and the patients were severely end-stage when treated – with some being on the transplant waiting list. In COPD, recent trials have shown NIV may be of benefit in stable hypercapnic patients, as discussed further in the section entitled “Chronic NIV in COPD”. In Duchenne patients, median survival is now nearly 30 years, and around a third of our Duchenne patients are living into their late thirties and early forties. NIV has been extended to the paediatric age range, with the feasibility of using NIV to control nocturnal hypoventilation in children initially being demonstrated Polio TB Neuromuscular

100

Continuing NIV %

80

Kyphoscoliosis

60 COPD

40 20 Bronchiectasis 0

0

1

2

3 Years

4

5

6

Figure 4.  Probability of continuing NIV long term, which is equivalent to survival in most cases. Reproduced from Simonds et al. (1995) with permission from the publisher. ERS Practical Handbook Noninvasive Ventilation

5

Introduction

predominantly in neuromuscular conditions. Many of these children now survive to adolescence or adulthood, as shown in the section entitled “Chronic NIV in hereditary neuromuscular disorders”. Furthermore, characterisation of the genotypes and phenotypes of some congenital disorders (e.g. congenital myasthenia and congenital muscular dystrophies) has clarified the natural history of these conditions, thereby facilitating anticipatory care plans and enabling personalised ventilatory care. Present trends There is also growing interest in NIV in cardiology. There is no doubt that patients with heart disease and OSA benefit from treatment of the OSA. By contrast, Cheyne-Stokes respiration is a form of central sleep apnoea that has been recognised for centuries in chronic heart failure. It was previously thought to be simply a marker of severe disease and an epiphenomenon, but recently, the link to the progression of disease has been explored and it has been found to be more prevalent in milder cases. This is important, as heart failure is common and the majority of those affected have mild cardiac impairment. Recent work shows that around half of patients with mild heart failure have sleep disordered breathing too. CPAP can be used to treat OSA but does not work in central sleep apnoea or Cheyne–Stokes respiration. The new ventilatory concept of ASV aims to smooth out the Cheyne–Stokes pattern and, in doing so, may reduce associated sympathetic stimulation and arousals from sleep (see the section entitled “Chronic NIV in heart failure patients: ASV, NIV and CPAP”). Several large international multicentre trials of ASV in heart failure patients with predominant central sleep apnoea are now in progress, with cardiac and all-cause mortality or unplanned admissions as major outcome measures. However early results suggest ASV may not improve outcome contrary to expectations, and may cause harm in severe heart failure patients with central sleep apnoea. Palliative care NIV is now being used in some situations to palliate symptoms without the aim of prolonging survival or substantially modifying arterial blood gas tensions. Here, goals such as reduction in dyspnoea and control of symptoms of nocturnal hypoventilation should be set pre-emptively so that if these are not met, NIV can be discontinued and palliative efforts directed elsewhere. Nava et al. (2013) have shown that in oncology patients with solid tumours complicated by ARF and an expected life expectancy of 20 mbar) and a controlled form of NIV with higher settings of breathing frequency, aiming for normocapnia, and is needed to improve longterm survival, lung function, exercise tolerance and health-related quality of life in COPD patients. In line with this, the need for higher IPAP levels is often reported in patients suffering from OHS when targeting to approach normocapnia. As mentioned, this has to be titrated on the individual patient and their tolerance. Pressure-preset

Volume-preset

100

Ventilation %

80 60 40 20

All Austria Belgium Denmark Finland France Germany Greece Ireland Italy Netherlands Norway Poland Portugal Spain Sweden UK

0

Figure 2.  Percentages of pressure- and volume-preset positive pressure ventilators used for home mechanical ventilation in different countries. n=21 526. Reproduced from Lloyd-Owen et al. (2005), with permission from the publisher. ERS Practical Handbook Noninvasive Ventilation

21

The basics: equipment

The occurrence of OSA as a comorbidity suggests a need for higher EPAP to stabilise the upper airways. However, with a higher EPAP, an even higher IPAP is needed, in order to apply the same level of inspiratory pressure support to augment the respiratory system. This has to be remembered during treatment to ensure a sufficient alveolar ventilation and reduction of hypercapnic carbon dioxide tension levels, which is one of the major goals of NIV. Furthermore, NIV can be applied using an assist or controlled mode or by combining both settings in the assist-control mode. All these options have been used successfully. As mentioned, especially in COPD patients, high-intensity NIV including a high backup rate and the target of achieving controlled ventilation has been reported to give positive outcomes. These positive outcomes could not be achieved with low-intensity NIV in the assist mode. A controlled mode of NIV (assisted pressure-controlled ventilation (APCV or ASSPCV)) reduces the workload of respiratory muscles more effectively than is possible using the assist mode. However, the assist mode (PSV) is primarily used in the acute setting. Irrespective of the chosen mode, the interaction between the patient and the ventilator is of high impact and the patient’s tolerance to either setting has to be monitored and reflected. A harmony between the patient and the ventilator is needed and “autotriggering” (fig. 3) or “fighting against the ventilator” have to be monitored and minimised. Interfaces and circuits for NIV The interface is crucial for applying NIV successfully. Interfaces are discussed further in the section entitled “Choosing the interface”. The most common interfaces are nasal or oronasal masks, which both show advantages and disadvantages (table 2). In acute care, where patients suffer from severe dyspnoea, mouth breathing is common. Thus, an oronasal mask covering mouth and nose is the better interface

Pressure by Thoracic ventilator movement

Pressure by polygraphy

Patient–ventilator synchrony Patient–ventilator asynchrony: “auto-triggering”

Figure 3. Patient–ventilator asynchrony during mechanical ventilation after five regular breaths. Periodic breathing patterns by the subject (thoracic movements) are not detected by the ventilator and breathing frequency (pressure curves) is increased without patient effort (so-called “auto-triggering”). Reproduced and modified from Storre et al. (2014a), with permission from the publisher.

22

ERS Practical Handbook Noninvasive Ventilation

The basics: equipment

Table 2.  Comparison of oronasal# and nasal masks Aspect

Oronasal mask

Mouth leak Mouth breathing Dead space First choice interface Communication Eating and drinking Expectoration

Nasal mask

No

Yes

Possible

Decreases NIV quality

Higher

Low

Acute care

Chronic care

Reduced

Possible

No

Possible

No

Possible

Risk of aspiration

Elevated

Reduced

Risk of aerophagia

Elevated

Reduced

Claustrophobia

Elevated

Reduced

Lower

Higher

Comfort #:

sometimes also known as facial or face masks. Reproduced and modified from Storre et al. (2008), with permission from the publisher.

for these patients, providing sufficient NIV within a short period of time, which is essential for a good outcome. In addition to oronasal masks, total face masks can be used, which cover the complete face. Total face masks can be used to prevent pressure marks on the bridge of the nose. However, this is mostly at the cost of an increase in the amount of leaks, since sealing the interface is more challenging with a larger interface. If an oronasal mask is not tolerated, treatment can also be performed in acute ventilatory failure using a nasal mask. In chronic care and home mechanical NIV, a nasal mask is the preferred interface, and occasionally nasal pillows are used (table 2 and fig. 4). Surprisingly, despite the great importance of correct interface choice, there is little scientific evidence and few studies have focused on the choice of interface or systematically studied differences regarding the quality of NIV. Thus, the published recommendations are expert opinions rather than evidence-based medicine. Today, a broad variety of different commercial interfaces exists, which enhances the chance of making an appropriate choice for most patients. Nevertheless, due to varying facial anatomies, there are still some patients who need individually modelled masks. In patients with neuromuscular disorders and permanent daily application of NIV, mouthpiece ventilation is reported to be a successful tool. Besides the different options when choosing the right interface, there are varying options for the selection of the ventilatory circuits between the patient and the ventilator. In chronic care, single-limb circuits are most often used, which can be divided into circuits using an integrated active exhalation valve and those with a passive exhalation valve (or intentional leaks). These intentional leaks can be integrated either into the circuit itself or into the interface, and nowadays these are the circuits most often used in chronic care. This trend has been influenced more by the manufacturers than scientific evidence. Most devices manufactured in the past decade have become smaller and cheaper and have used the passive ERS Practical Handbook Noninvasive Ventilation

23

The basics: equipment a)

Nasal mask

Facial mask

Tracheostomy

100

Lung HMV users %

80 60 40 20 0

Thoracic HMV users %

b) 100 80 60 40 20 0

Neuromuscular HMV users %

c) 100 80 60 40 20

All Austria Belgium Denmark Finland France Germany Greece Ireland Italy Netherlands Norway Poland Portugal Spain Sweden UK

0

Figure 4.  Proportions of interfaces used for home mechanical ventilation (HMV) in different countries for users with a) lung, b) thoracic and c) neuromuscular diseases. No data were available for Polish lung disease users. n=21 526. Reproduced from Lloyd-Owen et al. (2005), with permission from the publisher.

24

ERS Practical Handbook Noninvasive Ventilation

The basics: equipment

exhalation valve circuits only. In these passive exhalation valve circuits, leakage is always present during expiration as well as inspiration, which is necessary for the elimination of carbon dioxide. However, this amount of leakage can be disadvantageous when patients are dependent on supplemental oxygen in addition to mechanical ventilation. In these cases, the low-pressure oxygen flow used in home care is added to the circuit and is partly lost due the large amount of intentional leakage. As a consequence, the FIO2 is reported to be substantially lower when compared with a circuit with an active exhalation valve, where the leakage in the circuit is only present during expiration. Consequently, using a circuit with an active exhalation valve is suggested when supplemental oxygen has to be added to NIV. For more details, see the section entitled “Supplemental oxygen and humidification”. Further reading • Bayarassou AH, et al. (2013). Common mistakes leading to NIV failure. Minerva Pneumol; 52: 39–53.

• Dreher M, et al. (2010). High-intensity versus low-intensity non-invasive ventilation in patients with stable hypercapnic COPD: a randomised crossover trial. Thorax; 65: 303–308.

• Elliott MW (2004). The interface: crucial for successful noninvasive ventilation. Eur Respir J; 23: 7–8.

• Köhnlein T, et al. (2014). Non-invasive positive pressure ventilation for the treatment of severe stable chronic obstructive pulmonary disease: a prospective, multicentre, randomised, controlled clinical trial. Lancet Respir Med; 2: 698–705.

• Lloyd-Owen SJ, et al. (2005). Patterns of home mechanical ventilation use in Europe: results from the Eurovent survey. Eur Respir J; 25: 1025–1031.

• Mehta S, et al. (2001). Noninvasive ventilation. Am J Respir Crit Care Med; 163: 540–577.

• Simonds AK, et al. (1995). Outcome of domiciliary nasal intermittent positive pressure ventilation in restrictive and obstructive disorders. Thorax; 50: 604–609.

• Storre JH, et al. (2008). Noninvasive mechanical ventilation in chronic respiratory failure: ventilators and interfaces. In: Muir JF, et al., eds. Noninvasive Ventilation. 2nd Edn. ERS Monogr; 41: 319–337.

• Storre JH, et al. (2014a). Monitoring des Beatmungspatienten [Monitoring of patients receiving mechanical ventilation]. Pneumologie; 68: 532–541.

• Storre JH, et al. (2014b). Oxygen supplementation in noninvasive home mechan­ ical ventilation: the crucial roles of CO2 exhalation systems and leakages. Respir Care; 59: 113–120.

• Windisch W, et al. (2012). Target volume settings for home mechanical ventilation: great progress or just a gadget? Thorax; 67: 663–665.

ERS Practical Handbook Noninvasive Ventilation

25

The basics: equipment

Choosing the interface

Anne-Kathrin Brill

Interfaces connect the ventilator via circuit tubing to the patient and thereby allow the delivery of pressurised air into the upper airways and subsequently into the lungs. Choosing an appropriate interface is essential for successful NIV. The interface has to provide a good seal and needs to be tolerated by the patient at the same time. Interfaces can broadly be classified into six categories: nasal masks, nasal pillows, oronasal masks, total face or full face masks, oral masks or mouthpieces, and the helmet (fig. 1). With the exception of mouthpieces and the helmet, most interfaces are made of a soft cushion and a mask frame. They are normally secured with a head frame, headgear, or straps with velcro, clips or hooks.

Key points • Interface choice depends on clinical setting, equipment availability, the patient’s individual characteristics and safety considerations. NIV services should have a range of different masks and accessories. • Fit a correctly sized mask and headgear to minimise leak and increase comfort. Ensure that the interface complies with the ventilator and circuit tubing. • All interfaces can be used to provide NIV successfully, but in the acute setting interfaces covering the mouth and nose are advantageous. • If the patient cannot adapt to NIV, changing the interface can be helpful but should not delay intubation in critically ill patients if NIV is clearly failing. • Patients using NIV for >12 h per day should have a spare mask and can benefit from having two alternative interface types to alternate with. 26

ERS Practical Handbook Noninvasive Ventilation

ERS Practical Handbook Noninvasive Ventilation

Nasal pillows Subtype of nasal mask, also nasal plugs or nasal slings Applied externally to the nares

Helmet Transparent hood with collar Covers the whole head and all or part of the neck, no contact with the head Has at least two ports Most helmets have an anti-asphyxia valve

Nasal mask# Covers the whole nose but not the mouth With or without forehead spacer

Oral masks and mouthpieces Placed between the patient’s lips Mouthpieces have various degrees of flexion and are held in place by a lip seal or the teeth Oral masks can also have headgear as a securing system

Figure 1. The six main interface types for NIV. #: available as a vented or non-vented version. Image of the human head by Patrick J. Lynch reproduced from Wikimedia Commons under CC BY 2.5 licence.

Full face mask# Also called total face mask, cephalic or integral mask Covers mouth, nose and eyes and seals around the perimeter of the face

Oronasal mask# Covers mouth and nose Special subtype: hybrid masks (a combination of nasal pillows and an oral mask) With or without forehead spacer

The basics: equipment

27

The basics: equipment

Today, we can choose from a broad range of commercially available interfaces and the different types, shapes, materials, sizes and accessories allow a suitable interface to be found for nearly every patient. Custom-made masks are another option and can be helpful for patients with a prominent facial anatomy, but in ARF there is usually not enough time for their realisation. Many aspects have to be considered when the interface is initially chosen (table 1). These include the overall setting of NIV, availability of interfaces, and technical and patient-related factors, as well as safety aspects, experience of staff and costs. All interfaces can be used to treat respiratory failure and there is not enough evidence in the literature to provide strong general recommendations towards a single type of interface in terms of tolerance or efficacy. Each interface type has advantages and disadvantages (table 2), and the most suitable interface will depend on the situation and the patient’s individual characteristics. The first aspects to consider when choosing the interface are technical compati­ bility, i.e. which ventilator and circuit is used, safety aspects, size of the interface and the overall setting in which NIV is started (table 3). NIV masks are provided as vented or non-vented masks. Vented masks have integrated holes in the mask frame or a swivel elbow to remove carbon dioxide and to prevent rebreathing. Non-vented masks do not have an opening and need a separate option for carbon dioxide removal from the circuit. It is important to be aware of which interface and circuit is used. Therefore, prior to the start of NIV the method of carbon dioxide removal should be identified, its patency assured, and compatibility of circuit and interface checked.

Table 1. Factors influencing interface selection in NIV Setting

Technical

Respiratory failure Availability  Acute Circuit used  Chronic  Vented Location of NIV  Non-vented  ICU/HDU CO2 rebreathing  Emergency  Location of  room  exhalation  Post-operative Ventilator used   recovery room   Respiratory ward  Leak  compensation   General ward  Ventilator Estimated length  settings of use Type of headgear  Short-term  Long-term

Patient-related

Others

Age

Safety  Vomiting  Quick-release  strap  Anti-asphyxia  valve#

Facial anatomy  Recent facial  surgery   Nasal patency  Dentures   Motor function Interface/ headgear size Skin conditions Claustrophobia Mouth breathing Eating/speaking Expectoration

Experience of staff Airborne disease  TB  Influenza Accessories  Dressings   Chin strap   Tube adaptors  Lining

The main factors that may influence interface selection in NIV are listed, but the list is not exhaustive. HDU: high-dependency unit. #: present in vented masks and most helmets.

28

ERS Practical Handbook Noninvasive Ventilation

The basics: equipment

Table 2. Characteristics of the different interfaces Oronasal mask

Total face mask

Helmet

Nasal mask

Nasal prongs

Mouth pieces

Acute setting

•

•

•

⚬

⚬

⚬

Use outside HDU/ICU

•

•

•

•

•

Chronic setting

•

•

•

•

•

•

•

•

•

•

•

Nasal patency required

•

•

Coughing and expectoration is easier

•

•

⚬

Less claustrophobic More likely to have leaks in the acute setting

Useful for prominent facial anatomy

•

High level of noise

•

•

•

•

•

•

No pressure on the nasal bridge

•

High gas flow required Chance of eye irritation

⚬

•

• •

•

Speaking is easier

⚬

⚬

•

•

•

•

HDU: high-dependency unit; closed circles: applicable to the interface; open circles: an alternative, but less common or less frequent option. Reproduced and modified from Brill (2014) with permission from the publisher.

Other safety considerations include: • the risk of asphyxiation in case of vomiting • the risk of asphyxiation in case of ventilator malfunction • the patient’s level of consciousness and ability to remove the mask autonomously • potential disease transmission With nasal masks and the helmet the risk in case of vomiting or ventilator malfunction is usually lower, but patients using masks covering their mouth and nose should either be able to remove the mask themselves, be able to call for immediate help, or receive closer monitoring. Vented masks covering the mouth and nose and most helmets are usually equipped with an anti-asphyxiation valve that opens in case of unexpected ventilator failure. These valves should never be obstructed intentionally. ERS Practical Handbook Noninvasive Ventilation

29

The basics: equipment

Table 3. Vented and non-vented interfaces and suitable respiratory circuits Interface

Suitable respiratory circuits

Vented  The interface has one or multiple little holes in the frame or on the swivel elbow allowing passive CO2 removal

Single-limb “vented” or “intentional leak” circuit: CO2 removal occurs through the orifices sited within the interface or the swivel elbow

Non-vented  The interface is completely closed

Double-limb “non-vented” circuit: CO2 removal occurs through the ventilator

 Requires an option for CO2 removal within the circuit or ventilator

Single-limb “non-vented” circuit: CO2 removal occurs through an active non-rebreathing expiratory valve at the distal end of the inspiratory circuit or at the end of a short expiratory limb Single-limb “vented” or “intentional leak” circuit: CO2 removal occurs through a whisper swivel or a hole orifice at the end of the circuit, or a plateau exhalation valve placed between circuit and mask

NIV is a droplet-generating procedure that may expose healthcare workers who stand close to patients on NIV to potentially infectious droplets. If NIV has to be applied to patients with highly transmittable airborne disease, the use of a nonvented mask covering the nose and mouth within a non-vented tubing circuit and filtering of the exhaled gases can reduce environmental spread. Interfaces are available as disposable and reusable versions. Disposable interfaces are mainly used for shorter episodes of NIV in the acute setting (e.g. in emergency rooms, ICUs or post-operative recovery rooms). A change to a mask for long-term use should be considered if NIV has to be prolonged, as masks for long-term NIV: • are often easier for patients to handle themselves • can be used for longer periods of time • often allow more equal distribution of pressure over the face Interface strategies differ in acute and chronic settings (fig. 2). In ARF the priority is to quickly achieve efficient ventilation with the patient being able to tolerate the interface. In this situation most patients are in respiratory distress and breathe at a high respiratory rate. Patients often start mouth breathing to bypass the higher resistance of the nose. This can result in mouth leaks and compromise the efficacy of ventilation if nasal masks are used. Only a few studies have compared different interfaces in ARF, but overall masks covering the nose and mouth were associated with a quicker improvement in blood gases and less NIV failure, while nasal masks were often better tolerated in the longer term (Kwok et al., 2003; Girault et al., 2009; Navalesi et al., 2000). Therefore, the recommendation is to use interfaces

30

ERS Practical Handbook Noninvasive Ventilation

ERS Practical Handbook Noninvasive Ventilation

Consider NIV failure

No

NIV success

Patient improvement

No

Consider additional alternative interface e.g. additional mouthpiece in NMD Account for patient’s needs and preferences

Consider custom-made mask

Try other types of interface and accessories

Figure 2. Interface strategies for NIV in adult patients with ARF and chronic respiratory failure (CRF). NMD: neuromuscular disease.

Consider switching to a smaller interface, nightly use only Account for patient’s needs and preferences

Try alternative interface

Yes

Patient still suitable for NIV?

Check for side-effects: Pressure marks Leak into the eyes Comfort and sleep Adjust if necessary

Adapted to NIV?

Adapted to NIV? Yes

Check for non-intentional leaks, asynchrony, ventilator settings, circuit; adjust if necessary

Check for non-intentional leaks, asynchrony, ventilator settings, circuit; adjust if necessary Yes

Start with nasal mask (or interface of patient’s choice; consider patient’s abilities, needs and comorbidities)

Start with oronasal mask

No

CRF

ARF

Check compatibility of interface, ventilator and circuit. Identify where exhalation occurs. Choose correct headgear.

The basics: equipment

31

The basics: equipment

covering the nose and mouth as the first choice in ARF and to try switching to a smaller interface once the patient is more stable. Mainly, oronasal masks are used in acute NIV. Full face masks or helmets can also be used in selected patients, but they are often a second choice because of higher costs, and the necessity of having experienced staff and a higher level of monitoring to use the helmet. In acute NIV there is often not very much time to try different masks and more than one change of interface type is often not possible. Slightly more stable patients might benefit from a second change of interface if it can help to increase tolerance and efficacy of NIV and thereby prevent intubation. However, trying different interfaces should not delay intubation in critically ill patients if NIV is clearly failing despite best efforts. In the chronic setting there is more time to choose the most appropriate interface for the patient, leading to more individual choices. Patients can usually try different interfaces and with the exception of the helmet all other interface types are used in long-term NIV for chronic respiratory failure. Smaller masks are often used in this setting and it is appropriate to start NIV with a nasal mask unless: • the nose is not patent • the patient cannot keep the mouth closed (e.g. due to muscular weakness) • the patient wishes to use a mask covering mouth and nose In addition to a good mask fit, it is also important that the interface meets the patient’s needs and does not interfere too much with the patient’s daily routine and sleeping habits. Masks should be easy to apply and remove by the patient or caregivers. Quick-release straps or headgear that can be placed with one hand can be an option if the patient has impaired motor function. In addition, some masks allow for wearing glasses more easily than others. Patients who need NIV for >12 h per day or who are at risk of pressure sores may benefit from having different interface types to alternate between. This allows for intermittent pressure relief of the skin and can give patients more flexibility with eating and speaking. Independent of the clinical situation and the selected interface type, interfaces should always be fitted in the correct size and with a correctly sized and suitable securing system, to improve fit and comfort and to reduce unpleasant side-effects such as air-leaks or pressure marks. The use of sizing gauges (usually provided by the manufacturer) is encouraged. Masks should not overlay the eyes or lips and, at least in the chronic setting, masks should be fitted in the situation in which the interface will be used (i.e. with the patient seated for daytime use, or lying down supine in his/her favourite sleeping position, where a cushion might be able to support or interfere with mask fit). Large air-leaks affect the efficacy of NIV, promote patient–ventilator asynchronies and should be corrected immediately by refitting or changing the interface. Most ventilators designed for NIV can compensate for small air-leaks and small leaks can mostly be tolerated if they do not irritate the patient (e.g. eyes or noise) or disturb sleep. Over-tightening of the mask straps should be avoided as it will cause discomfort and might lead to skin damage. Leaks can be reduced with the help of accessories, for example chin straps in the case of mouth leaks or use of lining and dermal dressings to improve the seal. The latter are also useful to protect the skin by reducing pressure and friction during NIV. For patients with prominent

32

ERS Practical Handbook Noninvasive Ventilation

The basics: equipment

anatomy or at risk for the development of pressure ulcers, masks with softer cushions can be used. An adjustable forehead spacer or multipoint headgears can also help to distribute the pressure more evenly over the face and reduce pressure on the bridge of the nose. Regular reassessments of the mask fit and side-effects are essential to provide comfortable and successful NIV. Breathing patterns change with age and recommendations for adults cannot be translated directly into paediatric NIV. Interface choice and special considerations for children are described in the section entitled “Acute NIV in children, including venti­lator and interface choice”.

Further reading • Brill AK (2014). How to avoid interface problems in acute noninvasive ventilation. Breathe; 10: 230–242.

• Carron M, et al. (2013). Complications of non-invasive ventilation techniques: a comprehensive qualitative review of randomized trials. Br J Anaesth; 110: 896–914.

• Crimi C, et al. (2010). A European survey of noninvasive ventilation practices. Eur Respir J; 36: 362–369.

• Fraticelli AT, et al. (2009) Physiological effects of different interfaces during non­ invasive ventilation for acute respiratory failure. Crit Care Med; 37: 939–945.

• Garuti G, et al. (2014). Open circuit mouthpiece ventilation: concise clinical review. Rev Port Pneumol; 20: 211–218.

• Girault C, et al. (2009). Interface strategy during noninvasive positive pressure ventilation for hypercapnic acute respiratory failure. Crit Care Med; 37: 124–131.

• Kwok H, et al. (2003). Controlled trial of oronasal versus nasal mask ventilation in the treatment of acute respiratory failure. Crit Care Med; 31: 468–473.

• Nava S (2013). Behind a mask: tricks, pitfalls, and prejudices for noninvasive ventilation. Respir Care; 58: 1367–1376.

• Nava S, et al. (2009). Interfaces and humidification for noninvasive mechanical ventilation. Respir Care; 54: 71–84.

• Navalesi P, et al. (2000). Physiologic evaluation of noninvasive mechanical ventilation delivered with three types of masks in patients with chronic hypercapnic respiratory failure. Crit Care Med; 28: 1785–1790.

• Schettino GP, et al. (2003). Position of exhalation port and mask design affect CO2 rebreathing during noninvasive positive pressure ventilation. Crit Care Med; 31: 2178–2182.

• Sferrazza Papa GF, et al. (2012). Recent advances in interfaces for non-­invasive ventilation: from bench studies to practical issues. Minerva Anestesiol; 78: 1146–1153.

ERS Practical Handbook Noninvasive Ventilation

33

The basics: equipment

Online resources • Escarrabill J. How do I choose the interface? ERS Course Noninvasive ventilation: basic concepts, Hanover 2014. www.ers-education.org/events/courses/ noninvasive-ventilation-basic-concepts,-hanover-2014.aspx

• Escarrabill J, et al. Interfaces used for NIV. ERS Course Noninvasive ventilation: basic concepts, Hanover 2014. www.ers-education.org/events/courses/noninvasiveventilation-basic-concepts,-hanover-2014.aspx

• Wijkstra P. Air leakage during NIV: How important, how to avoid, how to handle? ERS Course Noninvasive ventilation: basic concepts, Hanover 2014. www.­ ers-education.org/events/courses/noninvasive-ventilation-basic-concepts,hanover-2014.aspx

34

ERS Practical Handbook Noninvasive Ventilation

The basics: equipment

Supplemental oxygen and humidification

Sundeep Kaul and Anita K. Simonds Supplemental oxygen Patients receiving NIV for either ARF or chronic respiratory failure, irrespective of aetiology, often require additional oxygen therapy to increase PaO2 and, therefore, improve tissue oxygen delivery. The introduction of ventilator support via NIV may itself correct the hypoxia; however, in many cases, supplemental ­oxygen is required. Some ventilators that deliver NIV are equipped with an oxygen blender, thereby allowing the operator to dial in the FIO2 and titrate according to oxygen saturations (SpO2) or PaO2. However, portable bilevel pressure ventilators commonly used to deliver NIV do not typically have an oxygen control. These NIV ventilators entrain room air and, on many machines, oxygen enrichment requires supplemental oxygen to be fed proximally into the circuit or directly into the mask. Indications Supplemental oxygen is indicated in type 1 or 2 ARF and chronic respiratory failure requiring NIV.

Key points • Supplemental oxygen is often required by patients receiving NIV. • Adding oxygen via the mask yields greater oxygen delivery to the patient rather than altering inspiratory pressures. • Achieving higher oxygen tensions without reducing patient comfort is optimal. • Humidification during NIV is a standard of care. • Reduced humidification can lead to increased airway resistance and mucus plugging. • Appropriate humidification can reduce patient discomfort and improve NIV tolerance. • Different humidification systems are available and each should be considered according to the clinical scenario. ERS Practical Handbook Noninvasive Ventilation

35

The basics: equipment

Set-up and practical tips • Ensure the oxygen saturation monitor has good contact with patient’s fingertip and that the waveform is reliable (if the analogue signal is available). • Connect one end of the oxygen tube to the oxygen source (e.g. wall outlet or cylinder). • Connect the other end of oxygen tube to either the oxygen connector in the ventilator circuit or directly to the mask via an inlet or nipple. • Increase the oxygen flow rate (at the source) until the required SpO2 is achieved (typically above 92%). • Perform arterial blood gas analysis to confirm the readings. • Ensure that the ventilator is switched on and the oxygen supply is connected, as there have been fatal incidents due to unintended disconnections or failure to switch on the ventilator. The concentration of oxygen delivered to the patient depends on a number of factors, which include ventilator pressures and flow rates, mask leakage, the site of the exhalation port, oxygen flow rate, the site of oxygen entrainment, and patient’s inspiratory flow rate. If the clinical situation requires the amount of inspired oxygen to be increased swiftly, then manoeuvres available to the operator, in addition to increasing the oxygen flow rate, include optimising the site of oxygen entrainment and altering the inspiratory/expiratory pressures. Different entrainment sites are associated with the delivery of different inspired oxygen concentrations (fig. 1). Entraining supplemental oxygen directly through the mask yields higher SaO2 than adding it to other points in the ventilator circuit. Lower inspiratory pressures tend to be associated with higher oxygen concentrations but this may reduce the VT ­de­livered. High expiratory pressures may reduce tolerability. Humidification The normal, unsupported human airways are capable of delivering a VT of air at 100% relative humidity, at core temperature, when it reaches the alveolar surface 30 PO2 mmHg

25

▲

20

Site 1 Site 2 Site 3 Site 4

15 10

▲

5 0

▲ ▲ ▲

0

2

4 6 8 10 12 14 16 Oxygen flow rate L·min–1

Figure 1.  Effect of increasing flow rates on oxygen tension (PO2 ) at different entrainment sites at IPAP 10 cmH2O and EPAP 4 cmH2O. Supplemental oxygen was added at four sites along the ventilatory circuit. Site 1: between mask and exhalation port. Site 2: just distal to exhalation port. Site 3: at ventilator outlet, between ventilator circuit and ventilator. Site 4: directly into the mask via an inlet. Data from Kaul et al. (2006).

36

ERS Practical Handbook Noninvasive Ventilation

The basics: equipment

for optimal gas exchange. Humidification of inspired gas is typically provided by evaporation of water from tracheobronchial secretions. Evaporation results in airway cooling, which is partially compensated for by heat and water provided by the bronchial blood flow. During expiration, some of the moisture is recovered as the air flows back over the mucosa on its way out of the airways. Patients receiving NIV require sufficient humidification of inspired air for the same reasons as those who breathe spontaneously. Humidifying inspired gas to 100% relative humidity precludes heat and moisture exchange with the ­mucosa, and thereby excludes the need for recovery of heat and moisture from expired gas. NIV delivers air at higher inspiratory flow rates than the patient’s own ­re­spiratory flow rates, which can overwhelm the capacity of the patient’s ­humidification system. This, in turn, can lead to airway dryness, altered airway mechanics, inspissated secretions and increased work of breathing, and predispose to NIV failure. Humidification during NIV is an accepted standard of care. The aim of humidification during NIV is to reduce the risk of: • • • •

increased airway resistance mucus plugging of the airways ciliary dysmotility epithelial desquamation

Maintenance of normal mucosal function also sustains airway patency and lung compliance, thus preserving lung mechanics. Correct application of a humidification system may enhance patient comfort and reduce the work of breathing, thereby improving NIV compliance and success. When optimising the humidification process during NIV, it is important to take the following factors into account. Types of humidification systems There are two methods of providing humidification. Firstly, via heated humidifiers that utilise an external power source and water supply, and secondly, using a heat and moisture exchanger (HME) system, which recycles the patient’s own moisture and heat. Use the appropriate humidification system according to the patient’s needs. The pros and cons of each system are highlighted in table 1. Leaks Air leaks (via the mask or mouth) result in unidirectional nasal air flow and, therefore, reduced mucosal recovery of moisture during the expiratory phase, leading to dry nasal mucosa and increased nasal airway resistance. This, in turn, reduces NIV compliance and efficiency. Performance of HME is also reduced, as expired air is less moist. One should institute measures to reduce leakage and consider heated humidification systems. Interfaces Nasal masks are associated with mouth leaks and, therefore, reduced humidification and its consequences, as described above. Within a helmet interface, there is a pool of exhaled gas that will contribute to humidification. Consider substituting a nasal for an oronasal mask, especially in mouth-breathers, those with secretion management issues and in the acute setting. If using a ERS Practical Handbook Noninvasive Ventilation

37

The basics: equipment

Table 1.  Advantages and disadvantages of HME filter and heated humidifier (HH) s­ ystems during NIV System

Advantages

Disadvantages

HME

Cost effective Extended use in the ICU Eliminates circuit condensation (hygroscopic models are recommended) A booster system applied to a hydrophobic HME may preserve AH capacity when incoming gases are delivered at a temperature 60 kg⋅m−2 and in those patients who have a poor initial response to NIV. It is essential that such patients undergo a careful risk assessment and that NIV is administered in the most appropriate clinical setting, which is dependent on the balance between the risk of NIV success or failure. There are few data to support specific ventilator strategies in the management of acute decompensated OHS, but as is the case with all administration of NIV, an experienced multidisciplinary team is vital for its successful application. However, a requirement of NIV application is to control upper airways obstruction and overcome the respiratory muscle load and, as a consequence, a high EPAP and high inspiratory driving pressure are required. Specifically, the EPAP should be set to optimise respiratory mechanics by modifying lung volume and the operating pos­ ition on the pressure–volume curve as well as overcoming upper airways obstruction in order to reduce the work of breathing. IPAP should be titrated to ensure satisfactory chest wall excursion with supplementary oxygen entrained at the lowest level to achieve target oxyhaemoglobin saturations of between 88% and 92%. The inspiratory time should be set to between 1.2 s and 1.4 s in line with the patient’s restrictive lung defect and the backup rate should be used to ensure mandatory ventilation during sleep. A target backup rate of between 12 and 14 breaths⋅min−1 is recommended. The details of NIV titration are shown in figure 1. To deliver such noninvasive ventilatory support, an oronasal mask is the preferred interface in the acute setting to reduce mask leak and maximise ventilation. As with all NIV prescriptions, the interface should be individually sized and fitted by experienced staff to reduce the risk of nasal bridge ulceration during therapy and to optimise patient comfort. Treatment duration should be ongoing, with short breaks for oral medication, oral hygiene, hydration and nutrition, during the first ERS Practical Handbook Noninvasive Ventilation

57

58 Other settings Supplemental oxygen: If persistent hypoxia (SpO2 60 mmHg) Usual setting: 8–10 cmH2O

Assess for underlying cause(s) of acute decompensation# and treat appropriately

The basics: patient selection

ERS Practical Handbook Noninvasive Ventilation

The basics: patient selection

24–48 h of therapy. When clinical stability has been achieved, weaning of NIV support from 24 h per day to night-time only can be undertaken with continuous and then intermittent cardiorespiratory physiological monitoring. In patients without a pre-existing diagnosis of obesity-related respiratory failure, this needs to be confirmed and other causes of chronic respiratory failure excluded based on overnight limited respiratory polygraphy, lung function testing and imaging, as per standard practice. In particular, a coexistent diagnosis of COPD should be sought. It has been established that the hospital outcome of patients receiving NIV for treatment of acute decompensated respiratory failure is better in patients with obesity-related respiratory failure than in those with COPD. However, the longterm outcome is similar between these two patient groups if the patients with obesity-related respiratory failure are not established on domiciliary NIV treatment on discharge from hospital. This highlights the potential benefit of home mechanical ventilation for obese patients with chronic respiratory failure and, although these patients can be safely discharged in the short term following an acute episode of respiratory decompensation, an assessment for long-term CPAP or NIV treatment is required in the early post-discharge period. Clinical tips • The initial NIV interface should be an oronasal mask or a total face mask. • During the initial acute period, when day and night ventilation is required, ­settings can be manipulated to improve daytime tolerance. For example, lower levels of EPAP can be applied during the day to improve tolerance, as OSA does not occur outside of sleep, and the backup rate can be reduced or the spontan­ eous mode can be applied. • Optimisation of cardiorespiratory morbidity and treatment of cor pulmonale with aggressive diuresis is required concurrently with application of NIV. • Allow patients regular breaks from NIV as necessitated by the need for medication, meals and personal hygiene in order to prevent skin breakdown on the nasal bridge and promote adherence. • It is essential to correct upper airways obstruction with adequate EPAP and an optimal sleeping position to prevent incremental IPAP increases due to failure to deliver pressure to the lower airways rather than inadequate pressure support. Further reading • Carrillo A, et al. (2012). Noninvasive ventilation in acute hypercapnic respiratory failure caused by obesity hypoventilation and chronic obstructive pulmonary disease. Am J Respir Crit Care Med; 186: 1279–1285.

• Duarte AG, et al. (2007). Outcomes of morbidly obese patients requiring mechan­ ical ventilation for acute respiratory failure. Crit Care Med; 35: 732–737.

• Masa JF, et al. (2010). Non-invasive ventilation in acute and chronic respiratory failure secondary to obesity. In: Elliott M, et al., eds. Non-Invasive Ventilation and Weaning: Principles and Practice. 1st Edn. London, Hodder Arnold; pp. 408–417.

ERS Practical Handbook Noninvasive Ventilation

59

The basics: patient selection

Patients with acute-on-chronic hypercapnic respiratory failure due to non-COPD obstructive lung disease and interstitial disorders Marieke L. Duiverman and Peter J. Wijkstra Respiratory failure occurs when the respiratory system cannot maintain gas exchange, causing dysfunction of other organs and threatening life. Respiratory failure can primarily affect oxygenation, manifested by hypoxaemia, or ventilation, manifested by additional hypercapnia. Hypoxaemic respiratory failure can be classified into four pathophysiological mechanisms: • • • •

impaired diffusion shunt ventilation–perfusion mismatch hypoventilation

While the first three mechanisms lead primarily to hypoxaemia and not hypercapnia, hypoventilation leads to both. Hypercapnic respiratory failure caused by hypoventilation develops secondary to a decreased ventilatory drive or a load placed on the system that cannot be overcome by the respiratory muscles.

Key points • In acute-on-chronic respiratory failure, the worsened imbalance between the load placed on the system and the capacity of the respiratory muscles can no longer be compensated. • In obstructive lung diseases, this imbalance will soon lead to hypercapnia, while in interstitial lung diseases, as these are lung parenchymal diseases, hypoxaemia occurs primarily, with hypercapnia only at later stages. • The evidence for NIV in acute exacerbations of nonobstructive lung diseases is limited but in practice, it is often worth trying to prevent intubation. • The evidence for NIV in acute exacerbations of interstitial lung diseases is even more limited; it often fails, especially in idiopathic pulmonary fibrosis, but might be tried to avoid endotracheal intubation. 60

ERS Practical Handbook Noninvasive Ventilation

The basics: patient selection

In patients with pre-existing lung diseases, pathophysiological mechanisms may coexist, leading to variable degrees of hypoxaemic and hypercapnic respiratory failure. Respiratory failure can be chronic, when the respiratory failure develops over weeks to months, or acute, when there is a sudden deterioration over hours to days. ARF can occur without pre-existing respiratory failure or on top of chronic respiratory failure (CRF). In chronic hypercapnic respiratory failure, excess carbon dioxide is buffered through bicarbonate retention in the kidney. As a consequence, pH remains normal. However, when there is a sudden deterioration in patients’ baseline status, the kidney does not have time to retain enough bicarbonate to buffer the increasing amount of carbon dioxide and respiratory acidosis develops, being a typical sign of acute-on-chronic hypercapnic respiratory failure (fig. 1). In this section, we will focus on the situation where ARF develops on top of chronic hypercapnic respiratory failure in obstructive pulmonary disease other than COPD and in interstitial lung diseases (ILDs). In COPD, the use of NIV to treat acute-onchronic hypercapnic respiratory failure is proven and incorporated into guidelines. However, for non-COPD obstructive lung disease and interstitial disorders, the use of NIV is outside the current guidelines but seems to be used rather frequently. Acute-on-chronic respiratory failure in non-COPD obstructive lung diseases Why patients with CRF can deteriorate so rapidly seems to be a consequence of an imbalance between an increased load (caused by increased airway resistance and a hyperinflation-driven decreased lung compliance) and a decreased capacity of the respiratory system. Although one might expect that during this process, respiratory muscles become fatigued, the occurrence of respiratory failure seems to be a consequence of decreasing VT in order to prevent fatigue (fig. 2). When optimal settings are applied, NIV counterbalances several of these factors. First, the application of external PEEP to a level close to, but not above, intrinsic Respiratory failure

Lung failure

Pump failure Hypoventilation

Impaired diffusion

Hypoventilation Shunt

Ventilation–perfusion mismatch

(Central) depression of respiration

Imbalanced load and capacity of respiratory muscles Exacerbation Respiratory muscle weakness

Hypoxaemia

Hypercapnia

Figure 1.  Causes of respiratory failure. ERS Practical Handbook Noninvasive Ventilation

61

The basics: patient selection

1.0 Fatigue zone

tI/ttot

0.8 0.6 0.4 0.2 0

0

0.2

0.4 0.6 Pdi/Pdi,max

0.8

Figure 2.  Relationship between inspiration time (tI) as a fraction of total duty cycle (ttot) and force output of the diaphragm (Pdi/Pdi,max). In the COPD patient, in stable condition, this relationship lies close to the fatigue zone (triangle). This is in contrast to a healthy subject (circle). When an exacerbation occurs, more force has to be delivered by the diaphragm (arrow), and patients would enter the fatigue zone (rectangle). However, to prevent entering the fatigue zone, patients decrease their tI leading to a decrease in VT, leading to hypercapnia (oblique rectangle). Reproduced from Demedts M, et al., eds (1999), Longziekten [Lung Diseases], with permission from the publisher (Koninklijke Van Gorcum, Assen, The Netherlands).

PEEP helps to reduce the pressure gradient between the distal and central airways that is established during dynamic hyperinflation. By doing so, patients need less reduction in intrathoracic pressure and, thus, less inspiratory muscle work to initiate inspiration. The addition of (enough) pressure support helps to further unload the inspiratory muscles while it also augments ventilation. NIV has also been shown to have bronchodilatory effects in asthma. Finally, NIV might improve ventilation–­perfusion matching. None of these is provided with the application of oxygen therapy as the sole intervention. Asthma In asthma, overt respiratory failure is very rare, as the treatment of asthma has improved. Unfortunately, some patients do not adhere to their therapy, are not treated optimally or have severe exacerbations despite optimal therapy, so ventilatory failure in asthma does still occur. Acute ventilatory failure in asthma usually occurs only with severe respiratory distress. Patients hyperventilate until they become so fatigued that they can no longer sustain the hyperventilation and start to retain carbon dioxide. Combined respiratory and metabolic acidosis is often present at this stage. Therefore, patients should be monitored closely and ventilator assistance should not be delayed until patients reach this ‘crossover’ point. Despite aggressive medical management with supplemental oxygen and pharma­cotherapy, some patients fail to improve, and require ICU admission and mechanical ventilation. In this case, if the condition allows, NIV is preferable to

62

ERS Practical Handbook Noninvasive Ventilation

The basics: patient selection

endotracheal intubation (ETI) and invasive mechanical ventilation (IMV), as this carries the risk of complications such as ventilator-associated pneumonia and barotrauma. In asthma, there have been studies investigating NIV as a treatment of ARF. However, as this occurs so infrequently, studies have suffered from small numbers and a lack of meaningful clinical outcomes. Some studies have shown shorter length of hospital stay, reduced airflow limitation and improvement in ­respiratory rate. However, as results from these studies are controversial, NIV for the management of acute asthma is not recommended in guidelines. Nevertheless, it might be beneficial in a subgroup of patients in the ICU, under careful observation, and with careful consideration of ventilator type, mask choice, settings (inspiratory pressures that are too high may lead to barotraumas and too much PEEP may lead to increased lung hyperinflation) and nebuliser position. Bronchiolitis Obliterative bronchiolitis is a disease process characterised by subepithelial inflammatory and fibrotic narrowing of the bronchioles, and is caused by a range of medical conditions and exposures. Clinically, it leads to a disease presentation with progressive dyspnoea and nonproductive cough over a period of weeks to months, and abnormal lung function frequently characterised by an obstructive pattern. The natural history and prognosis of obliterative bronchiolitis is highly variable; it may lead to respiratory failure. In adults, there is no evidence to support the use of NIV, both as a treatment of CRF due to obliterative bronchiolitis and ARF. In children, acute bronchiolitis, usually due to (viral) infections, occurs more frequently and can sometimes develop into CRF due to the development of obliterative bronchiolitis. NIV can be effective in both these situations. Cystic fibrosis Although CF is a multisystem disease, the primary cause of death is respiratory failure. Inflammatory processes in the airway wall, mucus plugging and lung parenchyma destruction due to bronchiectasis lead to progressive airway obstruction, hyperinflation, ventilation–perfusion mismatch and, together with a decrease in respiratory muscle strength, eventually may cause (chronic) respiratory failure. During an exacerbation of the disease, inflammation worsens, and both airway obstruction and the amount of sputum increase, leading to detrimental effects on gas exchange and patient condition. Although initially it was stated that NIV should not be used routinely as a treatment for an exacerbation of CF or bronchiectasis because the excessive amount of sputum would limit its effectiveness, studies (although not RCTs) and several reports have shown that NIV can be of benefit in this situation. The excessive amount of sputum does not seem to be a major problem; indeed, two studies have shown that during an exacerbation, NIV might even aid in sputum mobilisation. It has been hypothesised that NIV is of benefit during sputum expectoration manoeuvres as it might prevent respira­ tory muscle fatigue and airway closure during prolonged expirations, leading to improvement of alveolar ventilation and better sputum mobilisation. In practice, NIV has become a therapy worth trying during a CF exacerbation with acuteon-chronic respiratory failure in order to prevent intubation, with its associated complications. Non-CF bronchiectasis Bronchiectasis might develop in association with several conditions other than CF (e.g. in pulmonary TB or infections with nontuberculous ERS Practical Handbook Noninvasive Ventilation

63

The basics: patient selection

mycobacteria, Kartagener’s syndrome, or due to prior pneumonia) or might occur without a known cause (idiopathic). These patients often die from bronchiectasis-­ related disease with or without ARF. Similar to CF, it is often thought that the excessive amounts of sputum characteristic of non-CF bronchiectasis would limit the use of NIV in these patients, especially during an acute exacerbation of the disease, which is often elicited by infection. However, NIV might be of use in selected cases with less severe ARF due to bronchiectasis exacerbations, in which it might prevent ETI and shorten the length of ICU stay. One has to bear in mind that NIV failure rates are high in more severe ARF. In a study by Phua et al. (2010), failure was caused by an inability of the NIV to correct gas exchange sufficiently and not due to problems with sputum handling. Recently, a case was presented with a patient with exacerbated non-CF bronchiectasis being successfully treated using a pump-assisted venovenous system for extracorporeal carbon dioxide removal as an alternative to ETI following NIV failure. Although this is only a case, alternatives to IMV are more than welcome, as mortality rates of up to 40% once on IMV have been reported. Interstitial lung diseases ILDs are a heterogeneous group of disorders of the lung parenchyma with a known cause (such as those associated with drugs or collagen vascular disease) or unknown cause (idiopathic pulmonary fibrosis (IPF) and interstitial idiopathic pneumonia other than IPF, such as nonspecific interstitial pneumonia). As the pulmonary interstitium, blood vessels and air spaces may be affected, diffusion impairment, shunting and ventilation–perfusion mismatching may lead to re­ spiratory failure. As a consequence, these diseases are characterised by breathlessness and increasing hypoxaemia. Patients may experience acute deterioration of their condition without any identifiable cause, histopathologically characterised by diffuse alveolar damage. Deterioration may also be caused by an infection, heart failure or thrombo­ embolism. In all cases of deterioration, severe hypoxaemia is the primary problem. At late stages, the lungs become so stiff that the respiratory muscles can no longer sustain the imposed load and relative hypoventilation may lead to hypercapnia on top of the usually worsened hypoxaemia. When ARF develops in ILDs, outcomes are very poor, especially in IPF. Use of conventional IMV by ETI does not seem to improve outcome, especially in IPF and especially when no reversible cause is present or no alternative, such as lung transplantation, is provided. Little is known about NIV during an exacerbation of ILDs. Theoretically, prevention of complications caused by IMV, such as ventilatorassociated pneumonia, may be of benefit. Some small observational studies have reported that NIV can be an effective treatment to prevent ETI in patients with less severe disease. Results seem to be better with non-IPF ILD. However, in more than half of cases, NIV is unsuccessful and, especially in these cases, mortality rates are equally high. In general, a more “lung-protective” ventilatory strategy with low inspiratory and expiratory pressures is advised, as the lungs are usually already very damaged. As the primary problem is usually hypoxaemia, high FIO2 must be provided to achieve adequate oxygenation. However, the optimal ventilatory strategies are unproven and numbers of study subjects are too small to make conclusions; consideration

64

ERS Practical Handbook Noninvasive Ventilation

The basics: patient selection

of ventilatory strategies is mainly on theoretical grounds. Overall, it seems that with progressive disease, adjustments made to the ventilator are often unsuccessful in correcting gas exchange sufficiently. Insufficient ventilation on NIV may warrant the institution of other treatment options, such as extracorporeal membrane oxygenation, especially once a reversible cause is expected or when lung transplantation is an option. However, NIV might be used as a palliative measure in patients with do-not-intubate orders to reduce dyspnoea sensation. However, little is known about quality of life during these last hours in patients on NIV compared with other attractive alternatives available to relieve dyspnoea once death is approaching. Further reading • Alzeer AH, et al. (2007). Survival of bronchiectatic patients with respiratory failure in ICU. BMC Pulm Med; 7: 17.

• Arcaro G, et al. (2014). The successful management of a patient with exacerbation of non-cystic fibrosis bronchiectasis and bilateral fibrothorax using a venovenous extracorporeal carbon dioxide removal system. Respir Care; 59: e197–e200.

• Barker AF, et al. (2014). Obliterative bronchiolitis. N Engl J Med; 370: 1820–1828. • Bellemare F, et al. (1982). Effect of pressure and timing of contraction on human diaphragm. J Appl Physiol; 53: 1190–1195.

• Bellemare F, et al. (1983). Force reserve of the diaphragm in patients with chronic obstructive pulmonary disease. J Appl Physiol; 55: 8–15.

• British Thoracic Society Standards of Care Committee (2002). Non-invasive ventilation in acute respiratory failure. Thorax; 57: 192–211.

• Carson KV, et al. (2014). Noninvasive ventilation in acute severe asthma: current evidence and future perspectives. Curr Opin Pulm Med; 20: 118–123.

• De Backer L, et al. (2011). The effects of long-term noninvasive ventilation in hypercapnic COPD patients: a randomized controlled pilot study. Int J Chron Obstruct Pulmon Dis; 6: 615–624.

• Ganesh A, et al. (2015). Use of noninvasive ventilation in adult patients with acute asthma exacerbation. Am J Ther [In press DOI: 10.1097/MJT.0000000000000184].

• Gifford AH (2014). Noninvasive ventilation as a palliative measure. Curr Opin Support Palliat Care; 8: 218–224.

• Holland AE, et al. (2003). Non-invasive ventilation assists chest physiotherapy in adults with acute exacerbations of cystic fibrosis. Thorax; 58: 880–884.

• Lazner MR, et al. (2012). Non-invasive ventilation for severe bronchiolitis: analysis and evidence. Pediatr Pulmonol; 47: 909–916.

• Mallick S (2008). Outcome of patients with idiopathic pulmonary fibrosis (IPF) ventilated in intensive care unit. Respir Med; 102: 1355–1359.

• Mason RJ, et al. (2010). Murray and Nadel’s Textbook of Respiratory Medicine. 5th Edn. Philadelphia, Saunders. ERS Practical Handbook Noninvasive Ventilation

65

The basics: patient selection

• Mollica C , et al. (2010). Mechanical ventilation in patients with end-stage idiopathic pulmonary fibrosis. Respiration; 79: 209–215.

• Moran F, et al. (2013). Non-invasive ventilation for cystic fibrosis. Cochrane Database Syst Rev; 4: CD002769.

• Pallin M, et al. (2014). Noninvasive ventilation in acute asthma. J Crit Care; 29: 586–593.

• Papiris S, et al. (2002). Clinical review: severe asthma. Crit Care; 6: 30–44. • Phua J, et al. (2010). Noninvasive and invasive ventilation in acute respiratory failure associated with bronchiectasis. Intensive Care Med; 36: 638–647.

• Placidi G, et al. (2001). Short-term effects of positive airway pressure on sputum clearance by directed coughing: a crossover randomized study. Pediatr Pulmonol; 32: Suppl. 22, 313.

• Ram FS, et al. (2004). Non-invasive positive pressure ventilation for treatment of respiratory failure due to exacerbations of chronic obstructive pulmonary disease. Cochrane Database Syst Rev; 3: CD004104.

• Soma T, et al. (2008). A prospective and randomized study for improvement of acute asthma by non-invasive positive pressure ventilation (NPPV). Intern Med; 47: 493–501.

• Tomii K, et al. (2010). Role of non-invasive ventilation in managing life-threatening acute exacerbation of interstitial pneumonia. Intern Med; 49: 1341–1347.

• Vestbo J, et al. (2013). Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: GOLD executive summary. Am J Respir Crit Care Med; 187: 347–365.

• Vianello A, et al. (2014). Noninvasive ventilation in the event of acute respiratory failure in patients with idiopathic pulmonary fibrosis. J Crit Care; 29: 562–567.

• Westhoff M (2015). Akut-auf-chronisches respiratorisches Versagen bei interstitiellen Pneumonien. [Acute on chronic respiratory failure in interstitial pneumonias]. Med Klin Intensiv Med Notfmed; 110: 188–196.

66

ERS Practical Handbook Noninvasive Ventilation

The basics: patient selection

The patient with acute hypoxaemic respiratory failure excluding pulmonary oedema Pongdhep Theerawit, Yuda Sutherasan and Paolo Pelosi NIV has been used for patients with respiratory failure since the 1940s. Initially, NIV aimed to support ventilatory respiratory failure, e.g. during exacerbations of COPD. Later, the benefit of NIV was confirmed for cardiogenic pulmonary oedema. Nevertheless, the application of NIV in hypoxaemic respiratory failure is still controversial. In acute hypoxaemic respiratory failure (mostly patients with ARDS), the pathogenesis and physiological alterations are substantially different from those of cardiogenic pulmonary oedema. The alveolar damage, including injury of pulmonary capillary vessels, occurs in almost the entire lung, leading to fluid leakage into pulmonary tissues and marked inflammation. Although invasive mechanical ventilation is a standard treatment as a protective strategy in these patients, NIV can also play a role in acute hypoxaemic respiratory failure. The benefits of NIV, in terms of improvements in gas exchange abnormalities, protective effects for lung injury and improved mortality outcomes, need be addressed when selecting patients. Selecting suitable patients should be performed on an individual basis. The aims of this section are to describe the physiological effects of NIV in acute hypoxaemic respiratory failure, for alleviation of symptoms as well as for improvement of mortality outcomes. Physiological rationale for NIV in acute hypoxaemic respiratory failure Currently, ventilator strategies for protection of injured lungs are recommended by the standard guidelines for invasive mechanical ventilation. PEEP should be applied at a high enough level to reduce cyclic opening and closing of injured alveoli, whereas the V T should be low to prevent injury by overstretching. Although the use of neuromuscular blocking agents in early ARDS can improve the adjusted 90-day mortality and increase ventilator-free days, an animal model

Key points • NIV can improve oxygenation in acute hypoxaemic respiratory failure. • NIV relieves dyspnoea by reducing the work of breathing. • As there is a high rate of NIV failure for patients with acute hypoxaemic respiratory failure, it is very important that there is appropriate patient selection, intensive monitoring and prompt intubation if signs of NIV failure develop. ERS Practical Handbook Noninvasive Ventilation

67

The basics: patient selection

of acute hypoxaemic respiratory failure has demonstrated improvement of lung aeration during spontaneous breathing in the ventilator mode of airway pressure release ventilation compared with controlled ventilation (Wrigge et al., 2003). This is explained by the active diaphragmatic function in the synchronised ventilator setting, promoting alveolar recruitment in the dependent lung area. As a result of this, the synchronised spontaneous breathing provided by mechan­ical ventilation, including NIV, may be of benefit to acute hypoxaemic respiratory failure patients. NIV has also been demonstrated to reduce the work of breathing in nonsevere ARDS. After applying NIV, the patients feel more comfortable, as shown by a reduction in respiratory rate and parameters related to work of breathing. Improvement of oxygenation during NIV has been observed in ARF patients. This is caused by several mechanisms: • Modern NIV machines can provide a high concentration of oxygen, which is adjustable, as with the invasive mechanical ventilator. • The applied CPAP can work as PEEP in order to recruit and keep open the collapsed alveoli. This increases functional residual capacity and makes the lungs more uniform. • The CPAP decreases the gradient pressure across pulmonary capillaries, probably resulting in decreased extravascular lung water. In terms of inflammation, application of bilevel positive airway pressure ventilation can reduce the concentrations of inflammatory cytokines such as interleukin-8 measured in serum and bronchoalveolar lavage fluid. In addition, NIV minimises alveolar injury, as shown by the lung wet-to-dry weight ratio and the pathological score of lung injury, in an animal model with ARDS. NIV and outcomes in acute hypoxaemic respiratory failure NIV to avoid intubation in ARDS patients NIV has been used as the first-line ventilatory mode for ARDS in many centres. A multicentre survey by Antonelli et al. (2007) revealed that in ∼30% of 147 cases of ARDS, NIV was given as the first-line treatment. An important issue to note when using NIV as a firstline treatment for ARDS is the high failure rate. This failure rate may be as high as 50%, especially in patients with severe community-acquired pneumonia (CAP). Moreover, these patients in whom NIV fails have a very high mortality rate. However, providing NIV in ARDS patients can improve oxygenation, decrease work of breathing and decrease rate of ventilator-associated pneumonia. Thille et al. (2013) presented data from 113 acute hypoxaemic respiratory failure patients, with 82 cases of diagnosed ARDS. The rate of NIV failure in ARDS was 61% and increased according to the severity of ARDS. The mortality rate in the ICU was 25%. Among acute hypoxaemic respiratory failure patients, ARDS patients had a higher mortality rate than non-ARDS patients. Interestingly, the mortality rate of moderate-to-severe ARDS patients intubated after NIV failure was similar to those intubated without prior NIV (47% versus 44%). Additionally, no difference in time to intubation was observed between survivors and nonsurvivors in patients receiving NIV and requiring intubation within the first 96 h. NIV in pneumonia Applying NIV in severe pneumonia shows a high rate of failure. A prospective observational study conducted by Jolliet et al. (2001) enrolled

68

ERS Practical Handbook Noninvasive Ventilation

The basics: patient selection

24 severe CAP patients who required ICU admission. The definition of severe pneumonia included bilateral or multilobar involvement, a respiratory rate of >30 breaths⋅min−1, PaO2/FIO2 ratio 95%. It should be acknowledged that due to loss of surfactant and inflammation ventilation/perfusion mismatch may last up to 12–24 h after lavage. In general, we would continue NIV for at least 30 min after the end of bronchoscopy to allow recruitment of lung tissue. Support and FIO2 are tapered guided by SpO2. In some patients, support by NIV may be necessary for >12 h after bronchoscopy. Endoscopy  From practical perspective, bronchoscopy during NIV is not different from unassisted endoscopy. Both oral and nasal introduction of the bronchoscope are possible but, in general, oral introduction is easier due to the position of the orifice for the bronchoscope in the interface (figs 1 and 3). Adverse events Bronchoscopy in hypoxaemic patients should be considered a high-risk proced­ ure. Although NIV decreases the risks, in particular oxygen desaturation, clinicians should be well prepared for any adverse events related to bronchoscopy or NIV. Gastric hyperinflation increases the risk of aspiration and should be reduced by maintaining low inspiratory pressures, whenever possible. When hypoxaemia develops despite an increase in PEEP and FIO2, the procedure should be discontinued immediately. If severe or persistent hypoxaemia occurs, endotracheal intubation should be performed. Cardiac complications are rare and mostly secondary to the development of hypoxaemia. It should be acknowledged that clinically relevant complications are uncommon with appropriate patient selection. In our clinic, patients with rather severe hypoxaemia are deemed appropriate for bronchoscopy during NIV, but only under meticulous monitoring of peripheral oxygen saturation and heart rhythm, in the presence of a clinician trained for emergency endotracheal intubation. ERS Practical Handbook Noninvasive Ventilation

145

NIV and the ICU

Figure 3. Oral introduction of the flexible video-bronchoscope through the dedicated commercially available elbow.

A role for NIV in other endoscopic procedures? Clinical experience regarding use of NIV to facilitate endoscopic procedures other than bronchoscopy is rather limited. Very few studies have been reported in the literature. We will, therefore, discuss this issue in less detail. Severe orthopnoea in cardiac disease patients may increase the risks associated with transoesophageal ultrasound as this usually requires the patient to be in supine or lateral position. NIV has been used to facilitate transoesophageal ultrasound, for instance, during percutaneous aortic valve replacement. Due to the relatively large diameter of the ultrasound endoscope, the commercially available elbows (fig. 2) cannot be used. However, with a surgical cutter, a vertical hole can be made in the soft part of the interface, that allows leak free introduction of the endoscope. Very recently interfaces have been introduced that allow introduction of large diameter endoscopes (fig. 4a). Moreover, one type even allows application of NIV after the endoscope has been introduced (fig. 4b). In general, some level of sedation will be required during the procedure. We would prefer remifentanil due to its short half-life and favourable haemodynamic characteristics. Close monitoring of vital signs is required in these high-risk patients. Prolonged endoscopic procedures in gastroenterology, such as endoscopic retrograde cholangiopancreatography (ERCP), usually require the use of sedatives. This may increase the risk of pulmonary complications, in particular in patients with respiratory muscle weakness, COPD, obesity and heart failure. Despite the absence of highquality studies, the use of NIV during ERCP should be considered in these patients. Although some experts have advocated the use of nasal interfaces for NIV with ERCP,

146

ERS Practical Handbook Noninvasive Ventilation

NIV and the ICU

a)

b)

Connector for ventilator tubing

Sealed orifice for endoscope

Figure 4. a) Example of an interface with large orifice that allows the introduction of all diameters of endoscope, including those used with transoesophageal ultrasound or ERCP (VBM Endoscopy mask; VBM Medizintechnik, Sulz, Germany). b) Interface with a large orifice that can be applied after oral introduction of the endoscope (Janus ventilation mask; Biomedical Srl, Florence, Italy). Images used with permission of the suppliers.

the efficiency will be limited due to air leak with mouth opening. New types of interfaces allow introduction of the ERCP endoscope while using a full face mask (fig. 4). The patient’s vital signs should be monitored closely during the procedure. Likewise, NIV should be considered to facilitate percutaneous endoscopic gastrostomy in highrisk patients, in particular patients with severe respiratory muscle weakness such as ALS. Although the use of nasal masks has been advocated, the new full face masks (fig. 4) are probably superior, due to reduced air leak. However, as clinical experience is limited this should be performed only by professionals with considerable experience in both NIV and endoscopy, and with close monitoring of vital signs. Further reading • American Society of Anesthesiologists Task Force on Sedation and Analgesia by Non-Anesthesiologists (2002). Practice guidelines for sedation and analgesia by non-anesthesiologists. Anesthesiology; 96: 1004–1017.

• Ambrosino N, et al. (2011). Unusual applications of noninvasive ventilation. ­Eur Respir J; 38: 440–449.

• Cabrini L, et al. (2013). Non-invasive ventilation during upper endoscopies in adult patients. A systematic review. Minerva Anestesiol; 79: 683–694.

• Cracco C, et al. (2013). Safety of performing fiberoptic bronchoscopy in critically ill hypoxemic patients with acute respiratory failure. Intensive Care Med; 39: 45–52.

• Esquinas A, et al. (2013). Broncoscopia durante la ventilación mecánica no invasiva: revisión de técnicas y procedimientos [Bronchoscopy during non-invasive mechanical ventilation: a review of techniques and procedures]. Arch Bronconeumol; 49: 105–112.

• Heunks LM, et al. (2010). Non-invasive mechanical ventilation for diagnostic bronchoscopy using a new face mask: an observational feasibility study. Intensive Care Med; 36: 143–147.

ERS Practical Handbook Noninvasive Ventilation

147

NIV and the ICU

NIV in respiratory pandemics

Anita K. Simonds The definition of a pandemic is “an epidemic occurring worldwide, or over a very wide area, crossing international boundaries and usually affecting a large number of people”. Pandemics tend to be viral in nature, the most important recent patho­ gens being influenza virus, severe acute respiratory syndrome (SARS) coronavirus (CoV) and Middle Eastern respiratory syndrome (MERS)-CoV. Clearly, new pathogens may arise at any time with rapid spreading potential. The cornerstones of the development of a pandemic are that: • the causative organism emerges in humans • there is minimal or no population immunity • it spreads easily from person to person The latter prerequisite is met by airborne and droplet transmission as in influenza virus, SARS-CoV and MERS-CoV, or transmission of a highly infectious pathogen in bodily secretions, such as Ebola virus. There is likely to be crossover in that additional spread may occur by direct contact or fomites (objects capable of transmitting infection, such as clothing, furniture, door handles and toys). Important factors that determine the impact of the organism are its transmissibility, as indicated by the average reproduction number R (average number of people infected by a single infectious person), and severity, as measured by case fatality rate. This section will focus on airborne viruses, and the lessons learnt from respiratory management in SARS and H1N1 influenza, and will give pointers for patient care and infection control management in future outbreaks.

Key points • NIV is unsuitable in rapidly progressive acute lung injury, severe pneumonia and multiorgan failure in acute respiratory pandemics. • NIV may be helpful in early cases of acute respiratory infection (e.g. influenza) to avoid the need for intubation and invasive ventilation. • NIV generates droplets that are an infection control risk. Full personal protective equipment should be used by teams experienced in both infection control and the application of NIV. 148

ERS Practical Handbook Noninvasive Ventilation

NIV and the ICU

Potential role of NIV in respiratory pandemic infections Pneumonia and/or acute lung injury culminating in ARF are the most frequent serious complications of respiratory pathogens. In a heterogeneous group of patients with ARF treated with NIV, pneumonic consolidation was shown to be a poor prognostic factor, and in community-acquired pneumonia, NIV failure may be associated with higher mortality. However, subsequent studies and a metaanalysis suggested NIV is reasonable as first-line therapy in subgroups of patients with community-acquired pneumonia, particularly in those with COPD, although should not prevent the rapid escalation to intubation and invasive ventilation in those who are progressing rapidly or fail to respond. Furthermore, use of NIV in a patient with extensive consolidation/bilateral whiteout, profound hypoxaemia or multisystem failure is very unlikely to be successful and should be avoided. A multi­centre survey of NIV as a first-line intervention in ARDS in expert centres showed NIV use avoided intubation in 54% of patients, with a Simplified Acute Physiology Score II of >34 and an inability to improve FIO2/PaO2 after 1 h being key predictors of failure. It should be noted that patients with more than two system failures, haemodynamic instability and neurological disturbance were excluded from NIV. These considerations raise the possibility that NIV could be used in ­respiratory pandemics in limited situations shown in box 1. Experience of NIV in SARS SARS was first identified as being caused by coronavirus with high case severity in 2003. NIV use in cases in Hong Kong, mainland China and Canada has been reported predominantly in retrospective case series with NIV delivered by face mask. Firm conclusions are therefore difficult to draw as milder patients or those with single-organ failure were probably more likely to be allocated NIV and the studies are small. Yam et al. (2005) showed early application of NIV resulted in a reduction in mortality and need for endotracheal intubation, and Cheung et al. (2004) found that in early ARDS cases, NIV reduced the need for intubation, decreased ICU length of stay or enabled patients to avoid ICU admission. It should be noted, however, that in a Canadian series, 19% of patients became critically ill and for those requiring mechanical ventilation (invasive ventilation), mortality was 45%, with higher risk of death in patients >65 years of age or with comorbidities. Experience of NIV in H1N1 influenza Some of the lessons learnt from SARS were applied in the H1N1 influenza outbreak in 2009–2010. Here, up to 30% of hospitalised patients required admission to an ICU, with rapid and progressive respiratory failure. Although guidelines from most authorities, including the World Health Organization (WHO), US Centers for Disease Control and Prevention, and European Centre for Disease Prevention and Control, did not recommend NIV as first-line therapy, its role was explored in >20 published studies, which are summarised in a recent review (Esquinas et al., 2014). Box 1. Potential uses of NIV • Early use of NIV may reduce the need for intubation • NIV may be used to wean patients from invasive ventilation • NIV may be used for patients in whom it is a ceiling of ventilatory care and/or to palliate symptoms ERS Practical Handbook Noninvasive Ventilation

149

NIV and the ICU

The studies are disparate, include variable numbers of patients and only a small proportion were prospective. NIV failure rate ranged from 0% to 100% and mortality from 0% to 50%. The risk of NIV failure was increased in patients with refractory hypoxaemia and multiorgan failure. The most favourable outcome from NIV was seen in patients with: • low initial Acute Physiology and Chronic Health Evaluation (APACHE) and Sequential Organ Failure Assessment scores • no requirement for vasopressor • fewer radiological quadrants affected by consolidation In a Canadian group, 82% of patients were mechanically ventilated on the first day of ICU admission and of these, 33% received NIV. A high proportion of the NIV recipients (85%) progressed to invasive ventilation, again suggesting that in a rapidly deteriorating, severely ill patient group, invasive ventilation is the ventilatory management of choice. There are no systematic studies of the use of NIV to wean patients with H1N1 from invasive ventilation, but logic would suggest it to be of value here, particularly in those with comorbidities such as heart failure, neuromuscular disease, OHS and COPD. Pregnancy and H1N1 There are case reports of successful NIV use in pregnant patients with H1N1, and Zhang et al. (2012) reported a case series on almost 400 pregnant patients with H1N1. Just over 60% of women required critical care and 47% mechanical ventilation. Of the 186 who were treated with mechanical ventilation, 83 received NIV and in 38, this was successful. Mortality was lower in those who initially received NIV compared with those who were initially intubated (p=0.006) but on multivariate analysis, septic shock proved to be an important independent prognostic factor for NIV failure. Higher APACHE score, evidence of liver damage and central nervous system symptoms were associated with NIV failure compared with those with successful NIV outcome on univariate analysis. It should be noted that pregnant patients are a high-risk group overall and extreme caution is urged. NIV in MERS-CoV and novel coronaviruses There are a few anecdotal reports of CPAP/NIV use in patients with ARF due to MERS-CoV where mostly it failed. The WHO has produced interim guidance for management of novel coronaviruses (see Further reading), which is summarised below. • Recognise severe cases when severe respiratory distress may not be sufficiently treated by oxygen alone, even when delivered at a high flow rate (intractable hypoxaemia and high work of breathing). • Mechanical ventilation should be instituted early in patients with increased work of breathing or persistent hypoxaemia despite oxygen therapy. • Consider NIV if local expertise is available, when immunosuppression is present, or in mild ARDS without impaired consciousness or cardiovascular failure. • Use a low-volume, low-pressure lung protection strategy with a target VT of 6 mL⋅kg−1, plateau airway pressure 10 µm) that travel short distances because of their higher mass, and contaminate the bedside or local environment. Smaller droplets or aerosols will remain airborne for longer periods and disseminate over greater distances. An aerosol is usually characterised by droplets of