BMJ Open Resp Res: first published as 10.1136/bmjresp-2016-000170 on 15 May 2017. Downloaded from Guideline British Thoracic Society Guideline for oxygen use in adults in healthcare and emergency settings

BRO’Driscoll, L S Howard, J Earis, V Mak, on behalf of the BTS Emergency Oxygen Guideline Development Group

To cite: O’Driscoll BR, The full Guideline for oxygen use in adults fifth vital sign’ (supplemented by blood Howard LS, Earis J, et al. in healthcare and emergency settings, pub- gases when necessary), and the British Thoracic Society 1 Guideline for oxygen use in lished in Thorax provides an update to the inspired oxygen concentration should 2 adults in healthcare and 2008 BTS Emergency oxygen guideline. be recorded on the observation chart emergency settings . BMJ The following is a summary of the recom- with the oximetry result. (The other Open Resp Res 2017;4: mendations and good practice points. The vital signs are pulse rate, blood pres- e000170. doi:10.1136/ sections noted to within this summary refer sure, temperature and respiratory rate). bmjresp-2016-000170 to the full guideline sections. ○ Pulse oximetry must be available in all ▸ Additional material is locations where emergency oxygen is available. To view please visit EXECUTIVE SUMMARY used. Clinical assessment is recom- the journal (http://dx.doi.org/ ≥ 10.1136/bmjresp-2016- Philosophy of the guideline mended if the saturation falls by 3% or 000170) ✓ Oxygen is a treatment for hypoxaemia, below the target range for the patient. not breathlessness. Oxygen has not been ○ All critically ill patients outside of a The concise version of the proven to have any consistent effect on critical care area (e.g. intensive care

BTS Guideline should be read the sensation of breathlessness in non- unit (ICU), high dependency unit copyright. in conjunction with the full version: https://www.brit- hypoxaemic patients. (HDU), respiratory HDU), should be thoracic.org.uk/guidelines- ✓ The essence of this guideline can be assessed and monitored using a recog- and-quality-standards/ summarised simply as a requirement for nised physiological track and trigger emergency-oxygen-use-in- oxygen to be prescribed according to a system such as the National Early adult-patients-guideline/ target saturation range and for those Warning Score (NEWS). This guideline replaces the

2008 British Thoracic Society who administer oxygen therapy to 2. Target Oxygen prescription http://bmjopenrespres.bmj.com/ guideline for emergency monitor the patient and keep within the ○ Oxygen should be prescribed to oxygen use. target saturation range. achieve a target saturation of 94–98% ✓ The guideline recommends aiming to for most acutely ill patients or 88–92% achieve normal or near-normal oxygen or patient-specific target range for Accepted 17 November 2017 saturation for all acutely ill patients apart those at risk of hypercapnic respiratory from those at risk of hypercapnic respira- failure (tables 1–4). tory failure or those receiving terminal ○ Best practice is to prescribe a target palliative care. range for all hospital patients at the 1. Assessing patients time of admission so that appropriate ○ For critically ill patients, high concentra- oxygen therapy can be started in the tion oxygen should be administered event of unexpected clinical deterior- on September 24, 2021 by guest. Protected immediately (table 1 and figure 1)and ation with hypoxaemia and also to this should be recorded afterwards in ensure that the oximetry section of the patient’s health record. the early warning score (EWS) can be ○ Clinicians must bear in mind that sup- scored appropriately. plemental oxygen is given to improve ○ The target saturation should be written Department of Respiratory oxygenation, but it does not treat the (or ringed) on the drug chart or Medicine, Salford Royal underlying causes of hypoxaemia enteredinanelectronicprescribing Foundation NHS Trust, which must be diagnosed and treated system (guidance on figure 1). Salford, UK as a matter of urgency. 3. Oxygen administration ○ ○ Correspondence to The oxygen saturation should be Oxygen should be administered by Dr BR O’Driscoll; checked by pulse oximetry in all staff who are trained in oxygen ronan.o’[email protected] breathless and acutely ill patients, ‘the administration.

O’Driscoll BR, Howard LS, Earis J, et al. BMJ Open Resp Res 2017;4:e000170. doi:10.1136/bmjresp-2016-000170 1 BMJ Open Resp Res: first published as 10.1136/bmjresp-2016-000170 on 15 May 2017. Downloaded from Open Access

Table 1 Critical illnesses requiring high levels of supplemental oxygen (section 8.10) The initial oxygen therapy is a reservoir mask at 15 l/min pending the availability of reliable oximetry readings. For patients with spontaneous circulation and a reliable oximetry reading, it may quickly become possible to reduce the oxygen dose whilst maintaining a target saturation range of 94–98%. If oximetry is unavailable, continue to use a reservoir mask until definitive treatment is available. Patients with COPD and other risk factors for hypercapnia who develop critical illness should have the same initial target saturations as other critically ill patients pending the results of blood gas results after which these patients may need controlled oxygen therapy with target range 88–92% or supported ventilation if there is severe hypoxaemia and/or hypercapnia with respiratory acidosis. Additional comments Recommendations Cardiac arrest or resuscitation Refer to resuscitation guidelines for choice of delivery Recommendation E1 device during active resuscitation. Give highest possible inspired oxygen concentration during CPR until spontaneous circulation has been restored. Shock, sepsis, major trauma, drowning, Also give specific treatment for the underlying Recommendations anaphylaxis, major pulmonary haemorrhage, condition. E2–E4 status epilepticus Major head injury Early tracheal intubation and ventilation if comatose. Recommendation E5 Carbon monoxide poisoning Give as much oxygen as possible using a bag-valve Recommendation E6 mask or reservoir mask. Check carboxyhaemoglobin levels. A normal or high oximetry reading should be disregarded because saturation monitors cannot differentiate between carboxyhaemoglobin and oxyhaemoglobin, owing to their similar absorbances.

The blood gas PO2 will also be normal in these cases copyright. (despite the presence of tissue hypoxia).

COPD, chronic obstructive pulmonary disease; CPR, cardiopulmonary resuscitation; PO2, oxygen tension blood gases.

○ These staff should use appropriate devices and KEY CHANGES SINCE THE FIRST EDITION OF THIS flow rates in order to achieve the target saturation GUIDELINE PUBLISHED IN 2008 fi Methodology

range ( gure 2). http://bmjopenrespres.bmj.com/ ○ Staff should be trained in the use of a range of The evidence review methodology has changed from different oxygen delivery devices to ensure NICE methodology to the BTS NICE accredited guide- oxygen is delivered safely. line production process which is based on the Scottish 4. Monitoring and maintenance of target saturation Intercollegiate Guideline Network (SIGN) methodology ○ Oxygen saturation and delivery system (including and adheres to AGREE methodology (see section 1). flow rate) should be recorded on the patient’s monitoring chart. Evidence levels and grade of recommendation ○ Oxygen delivery devices and flow rates should be These are now in SIGN format (see section 1 and tables 6 adjusted to keep the oxygen saturation in the and 7). target range. Prompt clinical assessment is required if oxygen therapy needs to be initiated Evidence base on September 24, 2021 by guest. Protected or increased due to a falling saturation level. The evidence base for the guideline has been updated ○ Oxygen should be prescribed and a signature to August 2013 (and extended to mid-2016 for key refer- should be entered on the drug chart on each ences). None of the 2008 recommendations have been drug round. challenged by new evidence, but many of the existing 5. Weaning and discontinuation of oxygen therapy recommendations are supported by new information. ○ Oxygen should be reduced in stable patients with There have been many observational studies but few ran- satisfactory oxygen saturation. domised trials directly relevant to the guideline since ○ Oxygen should be discontinued once the patient 2008. can maintain saturation within or above the target range breathing air but the prescription for a target The remit of the guideline has been extended range should be left in place in case of future The new guideline covers not just emergency oxygen deterioration and to guide early warning scores use but most oxygen use in healthcare settings. It also (EWS/NEWS). covers short-term oxygen use by healthcare workers

2 O’Driscoll BR, Howard LS, Earis J, et al. BMJ Open Resp Res 2017;4:e000170. doi:10.1136/bmjresp-2016-000170 BMJ Open Resp Res: first published as 10.1136/bmjresp-2016-000170 on 15 May 2017. Downloaded from Open Access copyright. http://bmjopenrespres.bmj.com/ on September 24, 2021 by guest. Protected

Figure 1 Oxygen prescription guidance for acutely hypoxaemic patients in hospital. COPD, chronic obstructive pulmonary disease; ICU, intensive care unit; NIV, non-invasive ventilation; PO2, oxygen tension; PCO2, arterial or arteriolised carbon dioxide tension; SpO2, arterial oxygen saturation measured by pulse oximetry.

O’Driscoll BR, Howard LS, Earis J, et al. BMJ Open Resp Res 2017;4:e000170. doi:10.1136/bmjresp-2016-000170 3 BMJ Open Resp Res: first published as 10.1136/bmjresp-2016-000170 on 15 May 2017. Downloaded from Open Access

Table 2 Serious illnesses requiring moderate levels of supplemental oxygen if the patient is hypoxaemic (section 8.11) The initial oxygen therapy is nasal cannulae at 2–6 L/min (preferably) or simple face mask at 5–10 L/min unless stated otherwise. For patients not at risk of hypercapnic respiratory failure who have saturation below 85%, treatment should be started with a reservoir mask at 15 L/min and the recommended initial oxygen saturation target range is 94–98%. If oximetry is not available, give oxygen as above until oximetry or blood gas results are available. Change to reservoir mask if the desired saturation range cannot be maintained with nasal cannulae or simple face mask (and ensure that the patient is assessed by senior medical staff). If these patients have coexisting COPD or other risk factors for hypercapnic respiratory failure, aim at a saturation of 88–92% pending blood gas results but adjust to 94–98% if the PCO2 is normal (unless there is a history of previous hypercapnic respiratory failure requiring NIV or IMV) and recheck blood gases after 30–60 min, see table 4. Additional comments Recommendations

Acute hypoxaemia (cause not yet Reservoir mask at 15 L/min if initial SpO2 below 85%, Recommendations diagnosed) otherwise nasal cannulae or simple face mask. D1–D3 Patients requiring reservoir mask therapy need urgent clinical assessment by senior staff. Acute asthma Recommendations Pneumonia F1–F3 Lung cancer Deterioration of lung fibrosis or Reservoir mask at 15 L/min if initial SpO2 below 85%, Recommendation F4 other interstitial lung disease otherwise nasal cannulae or simple face mask Pneumothorax Needs aspiration or drainage if the patient is hypoxaemic. Most Recommendations patients with pneumothorax are not hypoxaemic and do not F5–F6 require oxygen therapy. Use a reservoir mask at 15 L/min if admitted for observation. Aim at 100% saturation (oxygen accelerates clearance of pneumothorax if drainage is not required). Pleural effusions Most patients with pleural effusions are not hypoxaemic. If Recommendation F7

hypoxaemic, treat by draining the effusion as well as giving copyright. oxygen therapy. Pulmonary embolism Most patients with minor pulmonary embolism are not Recommendation F8 hypoxaemic and do not require oxygen therapy. Acute heart failure Consider CPAP or NIV in cases of pulmonary oedema. Recommendations F9–F10 Severe anaemia The main issue is to correct the anaemia. Most anaemic Recommendations

patients do not require oxygen therapy. F11–12 http://bmjopenrespres.bmj.com/ Postoperative breathlessness Management depends on underlying cause. Recommendation J1 COPD, chronic obstructive pulmonary disease; CPAP, continuous positive airway pressure; IMV invasive mechanical ventilation; NIV, non-invasive ventilation; PCO2, arterial or arterialised carbon dioxide tension; SpO2, arterial oxygen saturation measured by pulse oximetry.

outside of healthcare settings, but domiciliary oxygen • Use of oxygen by voluntary rescue organisations and use by patients is covered by the BTS Guideline for other non-National Health Service first responders. – home oxygen use in adults.1 3 • High flow nasal cannulae (HFNC).

The scope of the guideline has been widened The structure and format of the guideline has been on September 24, 2021 by guest. Protected The present guideline includes the following new topics changed since 2008 and settings which have been requested by guideline users: The 2008 Guideline was published as a self-contained • Postoperative and perioperative care including document in Thorax.1 Additional educational materials patient-controlled analgesia. and other resources including audit tools were • Endoscopy and other procedures requiring conscious made available on the British Thoracic Society website. sedation. The new guideline exists in two complementary formats. • Palliative care settings including hospices. • A concise guideline which contains recommendations • Use of helium–oxygen mixtures (Heliox) and nitrous and good practice points is published here. oxide/oxygen mixtures (Entonox). • The full guideline including evidence review, physi- • Use of continuous positive airway pressure (CPAP). ology overview, illustrations and references is pub- • Use of oxygen by healthcare professionals in patients’ lished in Thorax1 and is available on the British homes. Thoracic Society website.

4 O’Driscoll BR, Howard LS, Earis J, et al. BMJ Open Resp Res 2017;4:e000170. doi:10.1136/bmjresp-2016-000170 BMJ Open Resp Res: first published as 10.1136/bmjresp-2016-000170 on 15 May 2017. Downloaded from Open Access

Table 3 Conditions for which patients should be monitored closely but oxygen therapy is not required unless the patient is hypoxaemic (section 8.13) If hypoxaemic, the initial oxygen therapy is nasal cannulae at 2–6 L/min or simple face mask at 5–10 L/min unless saturation is below 85% (use reservoir mask) or if at risk from hypercapnia (see below). The recommended initial target saturation range, unless stated otherwise, is 94–98%. If oximetry is not available, give oxygen as above until oximetry or blood gas results are available. If patients have COPD or other risk factors for hypercapnic respiratory failure, aim at a saturation of 88–92% pending blood gas results but adjust to 94–98% if the PCO2 is normal (unless there is a history of respiratory failure requiring NIV or IMV) and re-check blood gases after 30–60 min, see table 4. Additional comments Recommendations Myocardial infarction and acute coronary Most patients with acute coronary artery syndromes are Recommendation F13 syndromes not hypoxaemic and the benefits/harms of oxygen therapy are unknown in such cases. Unnecessary use of high concentration oxygen may increase infarct size. Stroke Most patients who had stroke are not hypoxaemic. Oxygen Recommendation F14 therapy may be harmful for non-hypoxaemic patients with mild-moderate strokes. Hyperventilation or dysfunctional Exclude organic illness. Patients with pure hyperventilation See section 8.13.3 breathing due to anxiety or panic attacks are unlikely to require oxygen therapy. Re-breathing from a paper bag may cause hypoxaemia and is not recommended. Most poisonings and drug overdoses (see Hypoxaemia is more likely with respiratory depressant Recommendation F15 table 1 for carbon monoxide poisoning) drugs, give antidote if available, for example, naloxone for opiate poisoning. Check blood gases to exclude hypercapnia if a respiratory depressant drug has been taken. Avoid high blood oxygen levels in cases of acid aspiration as there is theoretical

evidence that oxygen may be harmful in this condition. copyright. Monitor all potentially serious cases of poisoning in a level 2 or level 3 environment (high-dependency unit or intensive care unit) Poisoning with paraquat or bleomycin Patients with paraquat poisoning or bleomycin lung injury Recommendation F16 may be harmed by supplemental oxygen. Avoid oxygen unless the patient is hypoxaemic.

Target saturation is 85–88%. http://bmjopenrespres.bmj.com/ Metabolic and renal disorders Most do not need oxygen Recommendation F17 (Tachypnoea may be due to acidosis in these patients) Acute and subacute neurological and These patients may require ventilatory support and they Recommendation G4 muscular conditions producing muscle need careful monitoring which includes spirometry. If the weakness patient’s oxygen level falls below the target saturation, they need urgent blood gas measurements and are likely to need ventilatory support. Pregnancy and obstetric emergencies Oxygen therapy may be harmful to the fetus if the mother Recommendations is not hypoxaemic. H1–H4

COPD, chronic obstructive pulmonary disease; IMV, invasive mechanical ventilation; NIV, non-invasive ventilation; PCO2, arterial or arteriolised carbon dioxide tension, SpO2, arterial oxygen saturation measured by pulse oximetry. on September 24, 2021 by guest. Protected

INTRODUCTION conditions such as ischaemic heart disease or stroke Aim of the guideline based on custom and practice. About 34% of UK ambu- The key aim of this guideline is to make oxygen use in lance journeys in 2007 involved oxygen use.4 This trans- emergency and healthcare settings safer, simpler and lated to about two million instances of emergency more effective. Oxygen is probably the commonest drug oxygen use per annum by all UK ambulance services, used in the care of patients who present with medical with further use in patients’ homes, GP surgeries and in emergencies. Prior to the publication of the first British hospitals. Audits of oxygen use and oxygen prescription Thoracic Society Guideline for Emergency Oxygen Use have shown consistently poor performance in many in Adult Patients in 2008,1 ambulance teams and emer- countries, and most clinicians who deal with medical gency department teams were likely to give oxygen to emergencies have encountered adverse incidents and virtually all breathless or seriously ill patients and also to occasional deaths due to underuse and overuse of – a large number of non-hypoxaemic patients with oxygen.5 11

O’Driscoll BR, Howard LS, Earis J, et al. BMJ Open Resp Res 2017;4:e000170. doi:10.1136/bmjresp-2016-000170 5 BMJ Open Resp Res: first published as 10.1136/bmjresp-2016-000170 on 15 May 2017. Downloaded from Open Access

Table 4 COPD and other conditions requiring controlled or low-dose oxygen therapy (section 8.12) Prior to availability of blood gases, use a 24% Venturi mask at 2–3 L/min or 28% Venturi mask at 4 L/min or nasal cannulae at 1–2 L/min and aim for an oxygen saturation of 88–92% for patients with risk factors for hypercapnia but no prior history of respiratory acidosis. Adjust target range to 94–98% if the PCO2 is normal (unless there is a history of previous NIV or IMV) and recheck blood gases after 30–60 min. Additional comments Recommendations COPD and other conditions causing May need lower range if acidotic or if known to be Recommendations G1–G2 fixed airflow obstruction very sensitive to oxygen therapy. Ideally use ‘Alert and section 8.12.1 (eg bronchiectasis) cards’ to guide therapy based on previous blood gas results. Increase Venturi mask flow by up to 50% if respiratory rate is above 30 bpm Exacerbation of cystic fibrosis Admit to regional CF centre if possible, if not Recommendations G1, G3 discuss with regional centre or manage according to and G6 protocol agreed with regional CF centre. Ideally use ‘alert cards’ to guide therapy. Increase Venturi mask flow by by up to 50% if respiratory rate is above 30 bpm Neuromuscular disease, neurological May require ventilatory support. Recommendations G1, G4 condition and chest wall deformity Risk of hypercapnic respiratory failure and G6 Morbid obesity Recommendations G1, G5 and G6 CF, Cystic fibrosis; COPD, chronic obstructive pulmonary disease; CPAP, continuous positive airway pressure; IMV, invasive mechanical ventilation; NIV, non-invasive ventilation; PCO2, arterial or arteriolised carbon dioxide tension; SpO2, arterial oxygen saturation measured by pulse oximetry.

Table 5 Abbreviations for oxygen devices for use on bedside charts Add flow in L/min for example, N 2 L/min=nasal oxygen at 2 l/min. copyright. V28 4 l/min=28% Venturi mask at 4 l/min. RM 15 l/min=reservoir mask at 15 l/min. A, air N, nasal cannulae HFN, high flow nasal cannulae V24, Venturi mask 24% V28, Venturi mask 28% V35, Venturi mask 35% V40, Venturi mask 40% V60, Venturi mask 60% http://bmjopenrespres.bmj.com/ H28, humidified O2 28% H40, humidified O2 40% H60, humidified O2 60% RM, reservoir mask SM, simple face mask TM, tracheostomy mask CPAP, continuous positive airway pressure NIV, non-invasive ventilation

Historically, oxygen has been administered for three Against this background, the Standards of Care main indications of which only one is evidence-based. Committee of the British Thoracic Society (BTS) estab- First, oxygen is given to correct hypoxaemia because lished a working party in association with 21 other soci- severe hypoxaemia is clearly harmful to the human eties to produce an evidence-based guideline for body. Second, oxygen has been administered to ill emergency oxygen use in the UK. This led to the produc- patients in case they might become hypoxaemic. tion of the 2008 BTS Guideline for emergency oxygen on September 24, 2021 by guest. Protected Recent evidence suggests that, if impaired gas use in adult patients which was the world’s first guideline exchange does actually develop, this practice may actu- for emergency oxygen therapy.1 This guideline has been ally place patients at increased risk (see full Guideline implemented throughout the UK and in many other section 6.3). Third, a very high proportion of medical countries leading to over 500 citations in the medical lit- oxygen was administered because most clinicians erature up to the end of 2016. believed, prior to 2008, that oxygen can alleviate The purpose of the update to the 2008 guideline is to breathlessness in most circumstances. However, there is strengthen the evidence base of the previous guideline no good evidence that oxygen relieves breathlessness in based on revised methodology (which meets criteria non-hypoxaemic patients. There is evidence of lack of contained in the AGREE II Instrument) and to extend effectiveness or minimal effectiveness in mildly hypoxae- the evidence base to the end of 2013.12 Additionally, the mic breathless patients with chronic obstructive pulmon- remit of the 2008 Guideline has been broadened to ary disease (COPD) and advanced cancer (see full cover several new aspects of oxygen use and a broader Guideline sections 6 and 8.11.4). range of locations where oxygen might be used.

6 O’Driscoll BR, Howard LS, Earis J, et al. BMJ Open Resp Res 2017;4:e000170. doi:10.1136/bmjresp-2016-000170 BMJ Open Resp Res: first published as 10.1136/bmjresp-2016-000170 on 15 May 2017. Downloaded from Open Access copyright. http://bmjopenrespres.bmj.com/ on September 24, 2021 by guest. Protected

Figure 2 Flow chart for oxygen administration on general wards in hospitals. COPD, chronic obstructive pulmonary disease. EPR, electronic patient record; EWS, early warning score; NEWS, National Early Warning Score; SpO2, arterial oxygen saturation measured by pulse oximetry.

O’Driscoll BR, Howard LS, Earis J, et al. BMJ Open Resp Res 2017;4:e000170. doi:10.1136/bmjresp-2016-000170 7 BMJ Open Resp Res: first published as 10.1136/bmjresp-2016-000170 on 15 May 2017. Downloaded from Open Access

Table 6 SIGN evidence levels SIGN levels of evidence 1++ High-quality meta-analyses, systematic reviews of RCTs or RCTs with a very low risk of bias 1+ Well-conducted meta-analyses, systematic reviews or RCTs with a low risk of bias 1 Meta-analyses, systematic reviews or RCTs with a high risk of bias 2++ High-quality systematic reviews of case–control or cohort studies High-quality case–control or cohort studies with a very low risk of confounding or bias and a high probability that the relationship is causal 2+ Well-conducted case–control or cohort studies with a low risk of confounding or bias and a moderate probability that the relationship is causal 2− Case–control or cohort studies with a high risk of confounding or bias and a significant risk that the relationship is not causal 3 Non-analytic studies, eg, case reports, case series 4 Expert opinion RCT, randomised controlled trial.

Table 7 SIGN grades of recommendation Grades of recommendations A At least one meta-analysis, systematic review or RCT rated as 1++, and directly applicable to the target population; or A body of evidence consisting principally of studies rated as 1+, directly applicable to the target population, and demonstrating overall consistency of results B A body of evidence including studies rated as 2++, directly applicable to the target population, and demonstrating overall consistency of results; or Extrapolated evidence from studies rated as 1++ or 1+ C A body of evidence including studies rated as 2+, directly applicable to the target population and demonstrating overall consistency of results; or

Extrapolated evidence from studies rated as 2++ copyright. D Evidence level 3 or 4; or Extrapolated evidence from studies rated as 2+ Good practice points ✓ Recommended best practice based on the clinical experience of the guideline development group

Intended users of the guideline and target patient hospital setting and in other settings such as palliative http://bmjopenrespres.bmj.com/ populations care: This guideline is mainly intended for use by all health- • Critically ill or hypoxaemic patients. care professionals who may be involved in emergency • Hypoxaemic patients and patients at risk of hypoxaemia. oxygen use. This will include ambulance staff, first • Non-hypoxaemic patients who may benefitfrom responders, paramedics, doctors, nurses, midwives, phy- oxygen (eg, carbon monoxide poisoning). siotherapists, pharmacists and all other healthcare pro- fessionals who may deal with ill or breathless patients. Areas not covered by this guideline fi Advice is also provided for rst responders belonging to • Oxygen use in paediatrics: the present guideline voluntary organisations or other non-NHS bodies. applies only to patients aged >16 years. Information based on this guideline is available on the • Oxygen use for high altitude activities. on September 24, 2021 by guest. Protected BTS website for use in the following situations: • Oxygen use during air travel. • Hospital use. • Underwater diving and diving accidents. • Primary care use. • Oxygen use in animal experiments. • Ambulance use (supplemented by Ambulance service • Oxygen use in high-dependency units. 13 guidance based on this guideline). • Oxygen use in intensive care units. • Use by nursing staff and allied health professions. • Interhospital level 3 transfers. These abbreviated versions of the guideline contain • Hyperbaric oxygen. the key recommendations and tables and charts that are • Respiratory support techniques including tracheal relevant to the particular situation. intubation, invasive ventilation and non-invasive venti- lation (NIV) (CPAP is included). Areas covered by this guideline • Self-initiated use of oxygen by patients who have The guideline addresses the use of oxygen in three home oxygen for any reason. main categories of adult patients in the prehospital and • Ongoing care of hypoxaemic patients at home.

8 O’Driscoll BR, Howard LS, Earis J, et al. BMJ Open Resp Res 2017;4:e000170. doi:10.1136/bmjresp-2016-000170 BMJ Open Resp Res: first published as 10.1136/bmjresp-2016-000170 on 15 May 2017. Downloaded from Open Access

Limitations of the guideline a target saturation range of 88–92% is suggested This guideline is based on the best available evidence pending the availability of blood gas results (grade concerning oxygen therapy. However, a guideline can A for COPD, grade D for other conditions). never be a substitute for clinical judgement in individ- A4. Most non-hypoxaemic breathless patients do not ual cases. There may be cases where it is appropriate benefit from oxygen therapy, but a sudden reduc- for clinicians to act outwith the advice contained in this tion of ≥3% in a patient’s oxygen saturation within guideline because of the needs of individual patients, the target saturation range should prompt fuller especially those with complex or interacting disease assessment of the patient (and the oximeter signal) states. Furthermore, the responsibility for the care of because this may be the first evidence of an acute individual patients rests with the clinician in charge of illness (grade D). the patient’s care and the advice offered in this guide- A5 Since oxygenation is reduced in the supine position, line must, of necessity, be of a general nature and fully conscious hypoxaemic patients should ideally be should not be relied on as the only source of advice in allowed to maintain the most upright posture possible the treatment of individual patients. In particular, this (or the most comfortable posture for the patient) guideline gives very little advice about the management unless there are good reasons to immobilise the of the many medical conditions that may cause hypox- patient (eg, skeletal or spinal trauma) (grade D). aemia (apart from the specific issue of managing the B. Clinical and laboratory assessment of hypoxaemia and patients’ hypoxaemia). Readers are referred to other hypercapnia (see full Guideline section 7) guidelines for advice on the management of specific B1 Fully trained clinicians should assess all acutely ill conditions such as COPD, pneumonia, heart failure patients by measuring respiratory rate, pulse rate, etc. Some of these disease-specific guidelines may blood pressure, temperature and assessing circu- suggest slightly different approaches to emergency lating blood volume and anaemia. Expert assist- oxygen therapy whereas the present guideline aims to ance from specialists in intensive care or from provide simple all-embracing advice about oxygen other disciplines should be sought at an early therapy. stage if patients are thought to have major life- threatening illnesses and clinicians should be pre- Definitions of terms used in the guideline (and normal pared to call for assistance when necessary includ- values) ing a call for a 999 ambulance in prehospital care copyright. Terms such as hypoxaemia and normal values for or a call for the resuscitation team or intensive oxygen and carbon dioxide in blood gases are provided care unit (ICU) outreach team in hospital care in the full Guideline (sections 3–6). (grade D). B2. Oxygen saturation, ‘the fifth vital sign’,shouldbe Planned review and updating of the guideline checked by trained staff using pulse oximetry in all

The 2017 guideline will be reviewed by BTS within breathless and acutely ill patients (supplemented by http://bmjopenrespres.bmj.com/ 5 years from publication. blood gases when necessary) and the inspired oxygen device and flow rate should be recorded on the obser- Declarations of interest vation chart with the oximetry result (grade D). All members of the Guideline Group made declarations B3. Initial clinical assessment and subsequent monitor- ing of acutely unwell patients should include the of interest in line with the BTS Policy, and further ‘ ’ details can be obtained on request from BTS. use of a recognised physiological track and trigger system, such as the NEWS which may trigger clinical review due to hypoxaemia, need for supplementary SUMMARY OF GUIDELINE RECOMMENDATIONS AND oxygen or for other reasons (grade D). GOOD PRACTICE POINTS B4. For patients who are at risk of hypercapnic respira- A. Achieving desirable oxygen saturation ranges in acute illness tory failure, it is recommended that the relevant on September 24, 2021 by guest. Protected (see figures 1–2 and full Guideline sections 6 and 8) section of the 2017 NEWS chart should be used. A1. This guideline recommends aiming to achieve a Points are awarded if the oxygen saturation is below normal or near-normal oxygen saturation for all or above the target range (grade D). acutely ill patients apart from those at risk of hyper- Good practice points for clinical assessment of patients with sus- capnic respiratory failure (grade D). pected hypoxaemia A2. The recommended target saturation range for ✓ The medical history should be taken when possible in acutely ill patients not at risk of hypercapnic respira- an acutely breathless patient and may point to the diag- tory failure is 94–98% (grade D). nosis of a particular acute illness such as pneumonia or A3. For most patients with known COPD or other pulmonary embolism or an exacerbation of a chronic known risk factors for hypercapnic respiratory condition such as COPD, asthma or heart failure. failure (eg, morbid obesity, cystic fibrosis, chest wall ✓ Never discontinue oxygen therapy to obtain an oxim- deformities or neuromuscular disorders, or fixed etry measurement on room air in patients who airflow obstruction associated with bronchiectasis), clearly require oxygen therapy.

O’Driscoll BR, Howard LS, Earis J, et al. BMJ Open Resp Res 2017;4:e000170. doi:10.1136/bmjresp-2016-000170 9 BMJ Open Resp Res: first published as 10.1136/bmjresp-2016-000170 on 15 May 2017. Downloaded from Open Access

✓ Physical examination should be undertaken urgently. taken if there is any concern about the accuracy of This may provide evidence of a specific diagnosis a capillary sample (grade D). such as heart failure or a large pleural effusion, but it C2. Local anaesthesia should be used for all arterial is common for the cause of breathlessness to remain blood gas specimens except in emergencies (grade A). undiagnosed until the results of tests such as chest C3. Blood gases should be checked in the following radiographs are available. situations: ✓ Record arterial oxygen saturation measured by pulse ▸ All critically ill patients (Grade D). ▸ oximetry (SpO2) and consider blood gas assessment Unexpected or inappropriate fall in SpO2 below 94% in patients with unexplained confusion and agitation in patients breathing air or oxygen or any patient as this may be presenting feature of hypoaxaemia requiring oxygen to achieve the above target range. and/or hypercapnia (cyanosis is a difficult physical (Allowance should be made for transient dips in sat- sign to record confidently, especially in poor light or uration to 90% or less in normal individuals during with an anaemic or plethoric patient). sleep) (grade D). ✓ Carefully measure respiratory rate and heart rate ▸ Deteriorating oxygen saturation (fall of ≥3%) or increas- because tachypnoea and tachycardia are more ing breathlessness in a patient with previously stable common than a physical finding of cyanosis in hypox- chronic hypoxaemia (eg, severe COPD) (grade D). aemic patients. ▸ Most previously stable patients who deteriorate clinically ✓ ‘ Appropriate changes should be made to any track and require increased fraction of inspired oxygen (FiO2) and trigger’ system used to allow for a lower target to maintain a constant oxygen saturation (grade D). range in patients at risk of hypercapnic respiratory ▸ Any patient with risk factors for hypercapnic respira- failure. These patients should score no points for sat- tory failure who develops acute breathlessness, deteri- uration if within their target range and they should orating oxygen saturation, drowsiness or other score points if the oxygen saturation falls below the symptoms of carbon dioxide retention (grade D). target range or if the saturation rises above the target ▸ Patients with breathlessness who are thought to be at range while breathing oxygen. The 2017 update of risk of metabolic conditions such as diabetic ketoaci- the national NEWS chart has a special section for dosis or metabolic acidosis due to renal failure oximetry measurements for use with patients who (grade D).

have a target range 88–92% and it is recommended ▸ Any other evidence from the patient’s medical condi- copyright. that the 2017 NEWS chart should be used in all hospi- tion that would indicate that blood gas results would tals (see recommendation B4). be useful in the patient’s management (eg, an unex- ✓ ‘ ’ The presence of a normal SpO2 does not negate the pected change in track and trigger systems such as a need for blood gas measurements especially if the sudden rise of several units in the NEWS score or an patient is on supplemental oxygen therapy. Pulse unexpected fall in oxygen saturation of 3% or more,

oximetry will be normal in a patient with normal even if within the target range) (grade D). http://bmjopenrespres.bmj.com/ oxygen tension (PO2) but abnormal blood pH or Good practice point: patients requiring increased concentration carbon dioxide tension PCO2 or with a low blood of oxygen oxygen content due to anaemia. For this reason, ✓ The requirement for an increased concentration of blood gases and full blood count tests are required as oxygen is an indication for urgent clinical reassess- early as possible in all situations where these mea- ment of the patient (and repeat blood gas measure- surements may affect patient outcomes. ments in most instances, see recommendations W13 ✓ All clinical staff who use oximeters must be trained in and W18 for exceptions). their use and made aware of the limitations of oxim- D. Initial oxygen therapy; initial choice of equipment for etry. (Oximetry is a valuable clinical tool but subject patients who do not have critical illness (see figures 1–2 and to artefact and errors of interpretation). table 2 and full Guideline sections 8.9 and 10)i C. Arterial and capillary blood gases (see full Guideline sections Initial oxygen therapy in critical illness is covered in on September 24, 2021 by guest. Protected 7.1.3 and 8.4 and 8.5) the next section. C1. For critically ill patients or those with shock or D1. For acutely breathless patients not at risk of hyper- hypotension (systolic blood pressure <90 mm Hg), capnic respiratory failure who have saturations below the initial blood gas measurement should be 85%, treatment should be started with a reservoir obtained from an arterial sample. For most patients mask at 15 L/min in the first instance. The oxygen who require blood gas sampling, either arterial concentration can be adjusted downwards (using blood gases or arterialised earlobe blood gases may nasal cannulae at 1–6 L/min or a simple face mask be used to obtain an accurate measure of pH and at 5–10 L/min) to maintain a target saturation of – PCO2. However, the PO2 is less accurate in earlobe 94 98% once the patient has stabilised (grade D). blood gas samples (it underestimates the oxygen tension by 0.5– 1 kPa) so oximetry should be moni- tored carefully if earlobe blood gas specimens are iFor initial management of patients at risk of hypercapnic respiratory used and a repeat arterial specimen should be failure, see recommendations G1 and G2.

10 O’Driscoll BR, Howard LS, Earis J, et al. BMJ Open Resp Res 2017;4:e000170. doi:10.1136/bmjresp-2016-000170 BMJ Open Resp Res: first published as 10.1136/bmjresp-2016-000170 on 15 May 2017. Downloaded from Open Access

D2. In other cases of acute hypoxaemia without critical irrespective of the oximeter reading and PaO2 (grade illness or risk factors for hypercapnic respiratory D). failure, treatment should be started with nasal can- F. Oxygen therapy for specific conditions that frequently nulae (or a simple face mask if cannulae are not tol- require oxygen therapy (see tables 2 and 3 and full Guideline erated or not effective) with the flow rate adjusted to sections 8.11 and 8.13) achieve a saturation of 94–98% (grade D). Respiratory conditions with low risk of hypercapnic D3. If medium-concentration therapy with nasal cannu- respiratory failure: lae or a simple face mask does not achieve the F1. In acute asthma, aim at an oxygen saturation of 94– desired saturation, change to a reservoir mask and 98% (grade D). seek senior or specialist advice (grade D). F2. In cases of pneumonia who are not at risk of hyper- Good practice point capnic respiratory failure, aim at an oxygen satur- ✓ High flow nasal oxygen using specialised equipment ation of 94–98% (grade D). should be considered as an alternative to reservoir F3. In acute breathlessness due to lung cancer, aim at an mask treatment in patients with acute respiratory oxygen saturation of 94–98% unless there is coexist- failure without hypercapnia. For initial management ing COPD. See also ‘Oxygen use in palliative care’ of patients at risk of hypercapnic respiratory failure section (grade D). see recommendations G1 and G2. F4. In acute deterioration of pulmonary fibrosis or other inter- E. Oxygen therapy in critical illness (see table 1 and full stitial lung diseases, aim at an oxygen saturation of Guideline section 8.10) 94–98% or the highest possible if these targets E1. Use the highest feasible inspired oxygen for ventila- cannot be achieved (grade D). tion during cardiopulmonary resuscitation. Once spon- F5. In most cases of pneumothorax, aim at an oxygen sat- taneous circulation has returned and arterial blood uration of 94–98% if the patient is not at risk of oxygen saturation can be monitored reliably, aim hypercapnic respiratory failure (grade D). for a target saturation range of 94–98% and take an F6. In patients with pneumothorax having hospital arterial blood gas sample to guide on-going oxygen observation without drainage, the use of high con- therapy. If the blood gas shows hypercapnic respira- centration oxygen (15 L/min flow rate via reser- tory failure, reset the target range to 88–92% or voir mask) is recommended unless the patient is

consider mechanical ventilation (grade D). at risk of hypercapnic respiratory failure (grade copyright. E2. In critical illness, including major trauma, sepsis, shock D). and anaphylaxis, initiate treatment with a reservoir F7. In pleural effusion, aim at an oxygen saturation of mask at 15 L/min and aim at a saturation range of 94–98% (or 88–92% if the patient is at risk of hyper- 94–98%. This advice also applies to patients with capnic respiratory failure) (grade D). critical illness who have risk factors for hypercapnia F8. In pulmonary embolism, aim at an oxygen saturation – –

pending the results of blood gas measurements and of 94 98% (or 88 92% if the patient is at risk of http://bmjopenrespres.bmj.com/ expert assessment. In patients with spontaneous cir- hypercapnic respiratory failure) (grade D). culation and a reliable oximetry reading, it may be Non-respiratory conditions: possible to maintain a saturation of 94–98% using F9. In acute heart failure, aim at an oxygen saturation of lower concentrations of oxygen (grade D). 94–98% (or 88–92% if the patient is at risk of hyper- E3. In cases of drowning, aim at an oxygen saturation of capnic respiratory failure) (grade D). 94–98% once spontaneous circulation is restored F10. Continuous positive airway pressure (CPAP) with (grade D). entrained oxygen or high-flow humidified nasal E4. In patients with acute seizures due to epilepsy or other oxygen to maintain saturation 94-98% (or 88-92% if causes, high-concentration oxygen should be admi- at risk of hypercapnia) should be considered as an nistered until a satisfactory oximetry measurement adjunctive treatment to improve gas exchange in can be obtained and clinicians should then aim for patients with cardiogenic pulmonary oedema who on September 24, 2021 by guest. Protected an oxygen saturation of 94–98% or 88–92% if the are not responding to standard treatment (or non- patient is at risk of hypercapnic respiratory failure invasive ventilation (NIV) if there is co-existant hyper- (grade D). capnia and acidosis) (grade B). E5. In cases of major head injury, aim at an oxygen satur- F11. In anaemia, aim at an oxygen saturation of 94–98% ation of 94–98%. Initial treatment should involve or 88–92% if the patient is at risk of hypercapnic high-concentration oxygen from a reservoir mask at respiratory failure (grade D). 15 L/min pending availability of satisfactory blood Good practice point gas measurements or until the airway is secured by ✓ Correction of anaemia by blood transfusion should intubation (grade D). be based on national guidelines. E6. In cases of carbon monoxide poisoning, an apparen- F12. In sickle cell crisis and acute chest syndrome, aim for tly ‘normal’ oximetry reading may be produced an oxygen saturation of 94–98% or aim at the sat- by carboxyhaemoglobin, so aim at an oxygen satur- uration level that is usual for the individual patient ation of 100% and use a reservoir mask at 15 L/min (grade D).

O’Driscoll BR, Howard LS, Earis J, et al. BMJ Open Resp Res 2017;4:e000170. doi:10.1136/bmjresp-2016-000170 11 BMJ Open Resp Res: first published as 10.1136/bmjresp-2016-000170 on 15 May 2017. Downloaded from Open Access

Good practice point regarding sickle cell crisis from a reservoir mask and home oxygen should be ✓ Arterial or arterialised capillary blood gases should provided (grade D). be sampled if there is any doubt about the reliability G. Patients at risk of hypercapnic respiratory failure (see table 4 of oximetry during a sickle cell crisis. and full Guideline section 8.12) F13. In myocardial infarction and acute coronary syn- G1. (also A3) For most patients with known COPD or dromes, aim at an oxygen saturation of 94–98% or other known risk factors for hypercapnic respiratory 88–92% if the patient is at risk of hypercapnic failure (eg, morbid obesity, cystic fibrosis, chest wall respiratory failure (grade D). deformities or neuromuscular disorders or fixed F14. High concentrations of oxygen should be avoided airflow obstruction associated with bronchiectasis), in patients with stroke, unless required to maintain a target saturation range of 88–92% is suggested normal oxygen saturation. Aim at an oxygen satur- pending the availability of blood gas results (grade ation of 94–98% or 88–92% if the patient is at risk A for COPD, grade D for other conditions). of hypercapnic respiratory failure (grade D). G2. Some patients with COPD and other conditions are Good practice points regarding stroke management vulnerable to repeated episodes of hypercapnic ✓ Oxygen saturation should be monitored at least every respiratory failure. In these cases, it is recommended 4 hours throughout the day and night in patients that treatment should be based on the results of pre- with acute stroke and all episodes of hypoxaemia vious blood gas estimations during acute exacerba- treated. tions. For patients with prior hypercapnic failure ✓ Patients with hypoxaemia post-stroke require medical (requiring NIV or intermittent positive pressure ven- review to establish and treat the cause. tilation) who do not have an alert card, it is recom- ✓ Oxygen should only be given once the airway has mended that low-concentration oxygen treatment been cleared and at the lowest concentration neces- should be started using a 24% Venturi mask at 2– sary to achieve an oxygen saturation of 94–98% or 3 L/min (or a 28% Venturi mask at 4 L/min or 88–92% if the patient is at risk of hypercapnic respira- nasal cannulae at 1–2 L/min if a 24% mask is not tory failure. available) with an initial target saturation of 88–92% ✓ Oxygen should be given via nasal cannulae, unless pending urgent blood gas results. These patients there are clear indications for a different oxygen should be treated as a high priority by emergency

delivery system. services, and the oxygen concentration should be copyright. ✓ Patients with stroke and cardiorespiratory comorbid- reduced if the saturation exceeds 92% but increased ities should be positioned as upright as possible, in a if it falls below 88% (grade D). chair if possible (see recommendation A5). Good practice points for COPD and other conditions that ✓ Patients with a reduced level of consciousness after may cause hypercapnic respiratory failure stroke should be nursed in the recovery position with Diagnosis of COPD or suspected exacerbation of COPD: ✓

the paralysed side lowest. If the diagnosis is unknown, patients over 50 years of http://bmjopenrespres.bmj.com/ Suspected hyperventilation: age who are long-term smokers with a history of Good practice points regarding patients with suspected chronic breathlessness on minor exertion such as hyperventilation walking on level ground and no other known cause ✓ Organic illness must be excluded before making a of breathlessness should be treated as having sus- diagnosis of hyperventilation. pected COPD for the purposes of this guideline. ✓ Patients with a definite diagnosis of hyperventilation ✓ Spirometry should be measured at least once during should have their oxygen saturation monitored. hospital admissions for suspected COPD (as per the Those with normal or high SpO2 do not require National Institute of Health and Care Excellence oxygen therapy. (NICE) COPD Guideline).15 Measurement of spirom- ✓ Rebreathing from a paper bag can be dangerous and etry may confirm (or exclude) a diagnosis of airflow is NOT advised as a treatment for hyperventilation. obstruction, and the FEV1 level is a useful indicator of on September 24, 2021 by guest. Protected F15. In most poisonings, aim at an oxygen saturation of disease severity in COPD. 94–98% unless the patient is at risk of hypercapnic Immediate management of patients with known or sus- respiratory failure (grade D). pected COPD: F16. In poisoning by paraquat and poisoning by bleomycin, ✓ If the saturation remains below 88% in prehospital give oxygen only if the saturation falls below 85% care despite a 28% Venturi mask, change to nasal and reduce or stop oxygen therapy if the satur- cannulae at 2– 6 L/min or a simple face mask at 5 L/ ation rises above 88% (grade D). min with target saturation of 88–92% and alert the F17. In most metabolic and renal disorders, aim at emergency department that the patient is to be an oxygen saturation of 94–98% unless the treated as a high priority. patient is at risk of hypercapnic respiratory failure ✓ Patients with a respiratory rate >30 bpm should have (grade D). the flow rate from Venturi masks set above the F18. For patients with cluster headaches, oxygen should minimum flow rate specified for the Venturi mask be administered using a flow of at least 12 L/min packaging to compensate for the patient’s increased

12 O’Driscoll BR, Howard LS, Earis J, et al. BMJ Open Resp Res 2017;4:e000170. doi:10.1136/bmjresp-2016-000170 BMJ Open Resp Res: first published as 10.1136/bmjresp-2016-000170 on 15 May 2017. Downloaded from Open Access

inspiratory flow. Increasing the oxygen flow rate into Good practice point a Venturi mask does not increase the concentration Management of hypercapnia or respiratory acidosis of oxygen which is delivered. due to excessive oxygen therapy (avoidance of life-threaten- ✓ Patients with a significant likelihood of severe COPD ing rebound hypoxaemia): or other illness that may cause hypercapnic respira- ✓ If a patient is suspected to have hypercapnic or tory failure should be triaged as very urgent on respiratory failure due to excessive oxygen therapy, arrival in hospital emergency departments and blood the oxygen therapy must be stepped down to the gases should be measured on arrival in hospital. lowest level required to maintain a saturation range ✓ Prior to availability of blood gas measurements, use a of 88–92%. This may be achieved using 28% or 24% 24% Venturi mask at 2–3 L/min or nasal cannulae at oxygen from a Venturi mask or 1–2 L/min via nasal 1–2 L/min or 28% Venturi mask at 4 L/min and aim cannulae depending on oxygen saturation and subse- for an oxygen saturation of 88–92% quent blood gas measurements. Initial hospital management of patients with exacerba- ✓ Sudden cessation of supplementary oxygen therapy tion of COPD: can cause life-threatening rebound hypoxaemia with ✓ Patients with exacerbations of COPD need careful a rapid fall in oxygen saturations below the starting monitoring for hypercapnic respiratory failure with oxygen saturation prior to the start of supplementary respiratory acidosis which may develop in the course oxygen therapy. of a hospital admission even if the initial blood gases G3. Initial oxygen treatment of cystic fibrosis exacerbations were satisfactory. should be similar to the initial oxygen treatment of ✓ Avoid excessive oxygen use in patients with COPD. COPD exacerbations with target saturation 88–92% The risk of respiratory acidosis in patients with hyper- (see sections 8.12.1–8.12.2) (grade D). capnic respiratory failure is increased if the PaO2 is G4. In the initial management of musculoskeletal and above 10.0 kPa due to previous excessive oxygen use. neurological disorders with acute respiratory fail- ✓ If following blood gas measurements the pH and ure, or acute-on-chronic respiratory failure, aim at – PCO2 are normal, aim for an oxygen saturation of an oxygen saturation of 88 92% and measure 94–98% unless there is a history of previous hyper- blood gases to determine if NIV will be required capnic respiratory failure requiring NIV or intermit- (grade D).

tent positive pressure ventilation or if the patient’s Good practice point regarding patients with neurological copyright. usual oxygen saturation when clinically stable is disorders below 94% (these patients should have a target range ✓ Patients with respiratory failure due to neurological of 88–92%). Blood gases should be repeated at disorders or muscle disease are at high risk of dying – 30 60 min to check for rising PCO2 or falling pH. and require urgent assessment to determine if they ✓ Recheck blood gases after 30–60 min (or if there is are likely to require non-invasive or invasive ventilator

evidence of clinical deterioration) for all patients support rather than oxygen therapy. Monitor these http://bmjopenrespres.bmj.com/ with COPD or other risk factors for hypercapnic patients with blood gases and regular spirometry ’ respiratory failure even if the initial PCO2 measure- (forced vital capacity). Patients wishes regarding this ment was normal. form of treatment should established as early as pos- ✓ ≥ If the PCO2 is raised but pH is 7.35 ([H+] sible in the course of the illness, ideally before an ≤45 nmol/L) and/or a high bicarbonate level acute episode has developed. (>28 mmol/L), the patient has probably got long- G5. Morbidly obese patients (body mass index (BMI) standing hypercapnia; maintain target range of 88– >40 kg/m2), even without evidence of coexistent 92% for these patients. Blood gases should be obstructive sleep apnoea are at risk of hypoventila- – repeated at 30 60 min to check for rising PCO2 or tion and should be given titrated oxygen to main- falling pH. tain a target saturation of 88–92% (grade D).

✓ on September 24, 2021 by guest. Protected If the patient is hypercapnic (PCO2 >6 kPa or 45 G6. NIV should be considered for hypercapnic patients mm Hg) and acidotic (pH <7.35 or [H+] >45 nmol/L), with COPD, CF, neuromuscular disorders or morbid start NIV with targeted oxygen therapy if respiratory obesity who are at risk of hypercapnic respiratory acidosis persists for more than 30 min after initial stand- failure if the pH is <7.35 or [H+] >45 nmol/L ard medical management. (grade D) (see BTS/ICS Guidelines for the ✓ For patients using Venturi masks, consider changing Ventilatory Management of Acute Hypercapnic from Venturi mask to nasal cannulae once the Respiratory Failure in Adults).16 patient has stabilised. H. Oxygen use during pregnancy (see full Guideline section 8.14) ✓ For patients who use long-term home oxygen H1. Women who suffer from major trauma, sepsis or (LTOT) for severe COPD, a senior clinician should acute illness during pregnancy should receive the consider setting a patient-specific target range if the same oxygen therapy as any other seriously ill standard range of 88–92% would require inappropri- patients, with a target oxygen saturation of 94–98%. ate adjustment of the patient’s usual oxygen therapy The same target range should be applied to women while the patient is in hospital. with hypoxaemia due to acute complications of

O’Driscoll BR, Howard LS, Earis J, et al. BMJ Open Resp Res 2017;4:e000170. doi:10.1136/bmjresp-2016-000170 13 BMJ Open Resp Res: first published as 10.1136/bmjresp-2016-000170 on 15 May 2017. Downloaded from Open Access

pregnancy (eg, collapse related to amniotic fluid J4. Complicated upper GI endoscopy or procedures in embolus, eclampsia or antepartum or postpartum patients with cardiorespiratory comorbidity are espe- haemorrhage) (grade D). cially likely to lead to hypoxaemia and may also lead H2. Women with underlying hypoxaemic conditions to hypercapnia, especially if the patient is heavily (eg, heart failure) should be given supplemental sedated. It is recommended that blood gases should oxygen during labour to achieve an oxygen satur- be measured if such patients should require prolonged ation of 94–98% unless they are at risk of hypercap- oxygen administration. The routine administration of nic respiratory failure (target range 88–92%) oxygen is not recommended as it may delay the recog- (grade D). nition of respiratory failure (grade D). H3. Pregnant women who are fully conscious with no J5. Constant clinical assessment of the patient is crucial cardiovascular compromise may be managed in the at all stages of conscious sedation procedures and sitting position or if lying down should use the full monitoring of capnography or transcutaneous left lateral position (grade D). carbon dioxide levels may be a useful adjunct to H4. Pregnant women above 20 weeks gestation (uterine identify early respiratory depression (grade D). fundus at or above the level of the umbilicus) who J6. During the recovery period after procedures requir- are at risk of developing associated cardiovascular com- ing conscious sedation, supplemental oxygen should promise (eg, trauma and vaginal bleeding) should be titrated to achieve target saturations of 94–98% in be positioned to avoid aortocaval compression by most patients and 88–92% in those at risk of hyper- using left lateral tilt, manual uterine displacement capnic respiratory failure (see 10.5.1) (grade D). or by placing them in a full left lateral position Good practice points related to oxygen use in perioperative care (grade D). ✓ A target saturation of 94–98% is recommended for H5. Women who are more than 20 weeks pregnant with most surgical patients except those at risk of hyper- evidence of hypoxaemia associated with reduced con- capnic respiratory failure when a range of 88–92% sciousness or those requiring respiratory or cardio- should be achieved. vascular support or cardiopulmonary resuscitation ✓ Pulse oximetry monitoring is recommended for post- should be managed with left lateral tilt or manual operative patients despite the lack of evidence from uterine displacement (ideally to the left) to improve randomised studies.

cardiac output and oxygen delivery (grade D). ✓ Patients using patient-controlled analgesia (PCA) should copyright. H6. The use of oxygen supplementation during intra- have two-hourly oximetry observations because of the uterine fetal resuscitation during labour was wide- risk of hypoxaemia. Oxygen should be administered spread in the past, but there is no evidence of to keep patients within the appropriate target satur- benefit. There is weak evidence of harm to the ation range. fetus if supplemental oxygen is given for long ✓ A target saturation of 94–98% is advised in most

periods during uncomplicated labour. Overall, the patients having PCA except those at risk of hypercap- http://bmjopenrespres.bmj.com/ use of oxygen during labour is only required when nic respiratory failure when a range of 88–92% there is evidence of maternal hypoxaemia (oxygen should be achieved. saturation <94%) (grade D). ✓ There is conflicting evidence concerning the balance J. Oxygen use in perioperative care and during procedures of potential benefits and risks of perioperative hyperoxae- requiring conscious sedation (see full Guideline sections 8.15– mia to reduce the risk of surgical site infection in elective 8.16 and 10.11) surgery and there is no evidence for this practice in J1. Hyperoxaemia is not recommended routinely in the patients having emergency surgical procedures. More perioperative and postoperative period to reduce the trials are required for specific procedures, and more incidence of postoperative nausea and vomiting information is required concerning long-term mortal- (grade D). ity risks to patients with cancer. In the meantime, J2. All procedures involving conscious sedation warrant oxygen should not be used for this indication outside on September 24, 2021 by guest. Protected routine continuous monitoring of oxygen saturation of clinical trials. via pulse oximetry prior and during the procedure, K.Oxygenuseinpalliativecare(see full Guideline section 8.17) and in the recovery period, particularly fibre-optic K1. Oxygen use in palliative care patients should be bronchoscopy and upper GI endoscopy where a reduc- restricted to patients with SpO2 consistently <90% fi tion in arterial oxygen saturation (SaO2)iscommon, or patients who report signi cant relief of breath- particularly with concurrent use of sedation (grade C). lessness from oxygen. In non-hypoxaemic patients, fi J3. Signi cant arterial oxygen desaturation (SpO2<90% opioids and non-pharmacological measures should or fall of 4% or more that is prolonged (>1 min be tried before oxygen (grade B). during endoscopy procedures) should be corrected K2. In general, there is no role for the monitoring of by supplemental oxygen with the aim of achieving oxygen saturation or PaO2 in comfort-focused care target oxygen saturations of 94–98%, or 88–92% in in the last few days of life. If the patient appears those at risk of hypercapnic respiratory failure comfortable, oxygen levels are irrelevant and (grade D). should not influence care (grade D).

14 O’Driscoll BR, Howard LS, Earis J, et al. BMJ Open Resp Res 2017;4:e000170. doi:10.1136/bmjresp-2016-000170 BMJ Open Resp Res: first published as 10.1136/bmjresp-2016-000170 on 15 May 2017. Downloaded from Open Access

Good practice points related to oxygen use in palliative care N2. CPAP with entrained oxygen to maintain saturation Oxygen therapy for the symptomatic relief of breath- 94–98% should be considered to be an adjunctive lessness in palliative care patients is more complex than treatment to improve gas exchange in patients with the simple correction of hypoxaemia. Consider the fol- cardiogenic pulmonary oedema who are not lowing issues: responding to standard treatment in hospital care ✓ Consider early involvement of palliative care specia- or in prehospital care (grade B). lists and physiotherapists. Good practice point, high-flow humidified nasal oxygen via ✓ As breathlessness is a multifactorial sensation—a com- nasal cannulae prehensive assessment of contributing factors (such ✓ High flow nasal oxygen should be considered as a as anxiety) should be carried out. potentially superior alternative to reservoir mask ✓ Low-dose opioids should be considered because they treatment in patients with acute respiratory failure are effective for the relief of breathlessness in pallia- without hypercapnia.ii tive care patients. P. Patients with tracheostomy or laryngectomy (see section 10.3) ✓ A trial of a handheld fan to help relieve breathless- P1. When oxygen is required by patients with prior ness is recommended prior to trial of oxygen. tracheostomy or laryngectomy, a tracheostomy mask ✓ Oxygen use has to be tailored to the individual and a (varying the flow as necessary) should achieve the formal assessment made of its efficacy for reducing desired oxygen saturation (tables 1–4). An alterna- breathlessness and improving quality of life for that tive delivery device, usually a T-piece device fitted person. directly to the tracheostomy tube, may be necessary ✓ Oxygen therapy should not be continued in the if the patient deteriorates (grade D). absence of patient benefit or where its disadvantages Q. Humidification of oxygen (see section 10.2) (eg, discomfort of masks or nasal cannulae, drying of Q1. Humidification is not required for the delivery of mucous membranes) outweigh any likely symptom- low-flow oxygen (mask or nasal cannulae) or for the atic benefit. short-term use of high-flow oxygen. It is not therefore L. Mixtures of oxygen with other gases (Heliox and Entonox) required in prehospital care. Pending the results of Use of helium–oxygen mixtures (Heliox) (see full clinical trials, it is reasonable to use humidified Guideline section 8.18): oxygen for patients who require high-flow oxygen

L1. There is insufficient evidence to support the use of systems for more than 24 hours or who report upper copyright. Heliox either as an inhaled gas or as the driving gas airway discomfort due to dryness (grade D). for nebuliser therapy in adult patients with acute Q2. In the emergency situation, humidified oxygen use exacerbations of asthma or acute exacerbations of can be confined to patients with tracheostomy or COPD (AECOPD) except as part of randomised an artificial airway although these patients can be clinical trials or in exceptional circumstances managed without humidification for short periods

(grade D). of time (eg, ambulance journeys) (grade D). http://bmjopenrespres.bmj.com/ L2. A therapeutic trial of Heliox is reasonable in Q3. Humidification may also be of benefit to patients patients with mechanical upper airway obstruction with viscous secretions causing difficulty with expec- or postoperative stridor (grade D). toration. This benefit can be achieved using nebu- L3. Heliox use for patients with asthma or COPD should lised normal saline (grade D). be considered only in clinical trials or in specialist Q4. Bubble bottles which allow a stream of oxygen to hands for severe exacerbations that are not responding bubble through a container of water should not be to standard treatment (and in COPD patients where used because there is no evidence of a clinically sig- there are contraindications to intubation) (grade D). nificant benefit but there is a risk of infection M. Use of nitrous oxide/oxygen mixtures (Entonox) for anal- (grade D). gesia (see full Guideline section 9.11) Good practice points related to humidified oxygen therapy M1. The use of Entonox gas mixture for analgesia ✓ Consider use of a large volume oxygen humidifier on September 24, 2021 by guest. Protected should be avoided if possible in patients at risk of device for patients requiring high-flow rates or longer hypercapnic respiratory failure (grade D). term oxygen, especially if sputum retention is a clin- N. CPAP and humidified high-flow nasal oxygen ical problem. Use of CPAP in the perioperative period and for pul- ✓ In the absence of an artificial airway, the decision monary oedema (see online section 8.19): to humidify supplemental oxygen needs to be N1. Patients with diagnosed sleep-disordered breathing made on an individual basis but this practice is not established on CPAP undergoing surgery should evidence-based. bring their machines with them and use them in R. Driving gas for nebulised treatments (see section 10.4) the preoperative and postoperative period. If R1. For patients with asthma, nebulisers should be adequate saturations are not achieved despite CPAP driven by piped oxygen or from an oxygen cylinder therapy, then assess for worsening ventilation with blood gases and oxygen should be entrained to achieve a saturation of 88–92% (grade D). iiThe prefix O is not in use for recommendations.

O’Driscoll BR, Howard LS, Earis J, et al. BMJ Open Resp Res 2017;4:e000170. doi:10.1136/bmjresp-2016-000170 15 BMJ Open Resp Res: first published as 10.1136/bmjresp-2016-000170 on 15 May 2017. Downloaded from Open Access

fitted with a high-flow regulator capable of deliver- appropriate oxygen target range for every patient ing a flow rate of >6 L/min. The patient should be under their care (grade D). changed back to his/her usual oxygen mask or can- Good practice points related to prescribing and administering nulae when nebuliser therapy is complete. If the oxygen therapy to patients cylinder does not produce this flow rate, an air- ✓ Oxygen should be prescribed on the drug chart or elec- driven nebuliser (with electrical compressor) tronic prescribing system using a target saturation range. should be used with supplemental oxygen by nasal ✓ Oxygen should be prescribed to a target saturation cannulae at 2–6 L/min to maintain an appropriate range rather than prescribing a fixed concentration oxygen saturation level (grade D). of oxygen or fraction of inspired oxygen (see recom- R2. When nebulised bronchodilators are given to mendations A1, A2, A4 and A5) patients with hypercapnic acidosis, they should be ✓ In most emergency situations, oxygen is given to given using an ultrasonic nebuliser or else a jet nebu- patients immediately without a formal prescription. liser driven by compressed air and, if necessary, sup- The lack of a prescription should never preclude plementary oxygen should be given concurrently by oxygen being given when needed in an emergency nasal cannulae to maintain an oxygen saturation of situation. However, a subsequent written record must 88–92%. The same precautions should be applied to be made of what oxygen therapy has been given to patients who are at risk of hypercapnic respiratory every patient in a similar manner to the recording of failure prior to the availability of blood gas results all other emergency treatment. and the oxygen saturation should be monitored con- ✓ If a patient has an oxygen alert card, initial oxygen tinuously during treatment. Once the nebulised therapy should be based on the guidance on the treatment is completed for patients at risk of hyper- card until the results of blood gases are available. capnic respiratory failure, their previous targeted T. Monitoring and adjusting oxygen therapy (see full oxygen therapy should be reinstituted (grade D). Guideline sections 9–11) Good practice points T1. Pulse oximetry must be available in all locations Do not allow hypoxaemia to occur while administering where emergency oxygen is being used by health- nebulised treatments. care professionals (see also the limitations of using ✓ For hypoxaemic patients, oxygen therapy should con- pulse oximetry section 7.1.2) (grade D).

tinue during nebulised treatments. T2. All documents which record oximetry measure- copyright. Driving gas for nebulised treatment in ambulances: ments or blood gas results should state whether the ✓ During treatment by ambulance staff, oxygen-driven patient is breathing air or a specified oxygen deliv- nebulisers should be used for patients with asthma ery device and flow rate using the abbreviations and may be used for patients with COPD in the shown in table 5 (grade D). absence of an air-driven compressor system. If oxygen T3. In all situations where repeated blood gas measure-

is used for patients with known COPD, its use should ments are required, they should be measured as http://bmjopenrespres.bmj.com/ be limited to 6 min. This will deliver most of the neb- soon as possible, usually within 30 min of any treat- ulised drug dose but limit the risk of hypercapnic ment change, to determine if the proposed target respiratory failure (section 10.4). Ambulance services saturations are appropriate. Consider the use of an are encouraged to explore the feasibility of introdu- indwelling arterial catheter if multiple samples are cing battery powered, air-driven nebulisers or port- likely to be required (grade D). able ultrasonic nebulisers. T4. Adjustments should only be made by registered staff S. Prescribing oxygen therapy (see full Guideline section 11) who have been trained to administer oxygen. If the S1. Every healthcare facility should have a standard oxygen saturation falls below the prespecified oxygen prescription document or, preferably, a range, the concentration of oxygen should be designated oxygen section on all drug prescribing increased; if the saturation rises above this range, cards or guided prescription of oxygen in electronic the oxygen concentration should be reduced. If the on September 24, 2021 by guest. Protected prescribing systems (grade D). monitoring of oxygen saturation is performed by S2. A prescription for oxygen should always be provided, unregistered staff (eg, healthcare assistants), there except in sudden illness when it must be started imme- must be a clear protocol in place which requires diately and documented retrospectively (grade D). that they should inform staff who are trained to S3. Doctors and other prescribers should prescribe administer oxygen if the oxygen saturation is above oxygen using a target saturation range (sections 8, 9 or below the target saturation (grade D). and 11) and sign the drug chart or electronic pre- Good practice points related to administration of oxygen therapy scribing order (grade D). ✓ For hypoxaemic patients, oxygen therapy should con- S4. An oxygen target saturation range should be pre- tinue during other treatments such as nebulised scribed for all patients who are admitted to hospital. therapy. Clinicians should assess the clinical status of This will ensure that every patient will receive appro- the patient prior to prescribing oxygen and the priate oxygen therapy if it should be required. It will patient’s condition should be reassessed frequently also ensure that all clinicians are aware of the during oxygen use (see recommendations B1–B3).

16 O’Driscoll BR, Howard LS, Earis J, et al. BMJ Open Resp Res 2017;4:e000170. doi:10.1136/bmjresp-2016-000170 BMJ Open Resp Res: first published as 10.1136/bmjresp-2016-000170 on 15 May 2017. Downloaded from Open Access

✓ The administering healthcare professional should should also be stopped if the patient has come to note the oxygen saturation before starting oxygen the end of a written protocol of timed oxygen (eg, therapy wherever possible but never discontinue or postoperatively) (grade D). delay oxygen therapy for seriously ill patients (see U5. Oxygen saturation on air should be monitored for recommendation B2). 5 min after stopping oxygen therapy. If it remains ✓ The healthcare professional should start oxygen in the desired range, it should be rechecked at therapy using an appropriate delivery system and flow 1 hour (grade D). rate as specified in sections 8–10 of this guideline. U6. If the oxygen saturation and physiological ‘track The target oxygen saturation should be documented and trigger’ score (eg, NEWS) is satisfactory at on the respiratory section of the observation chart. 1 hour, the patient has safely discontinued oxygen ✓ Whenever possible, patients should be given an therapy. However, saturation and physiology should oxygen information sheet (example in web appendix continue to be monitored on a regular basis 6 on the BTS website). according to the patient’s underlying clinical condi- ✓ Staff should check the oxygen supply and connec- tion (grade D). tions on a regular basis because there have been U7. If the saturation falls below the patient’s target range serious incidents due to disconnection or misconnec- on stopping oxygen therapy, restart the lowest con- tion of oxygen supplies. centration that maintained the patient in the target ✓ Staff must ensure that adequate oxygen is provided range and monitor for 5 min. If this restores the sat- during transfers and while patients are in diagnostic uration into the target range, continue oxygen departments. Additionally, oxygen saturation should therapy at this level and attempt discontinuation of be monitored continuously for seriously ill patients oxygen therapy again at a later date provided the who require escorted transfers. This is because there patient remains clinically stable (grade D). have been serious incidents involving accidental dis- U8. If a patient requires oxygen therapy to be restarted continuation of oxygen or cylinders running out at a higher concentration than before to maintain during interward transfers or transfers to other the same target saturation range, the patient departments such as for X-rays. should have a clinical review to establish the cause U. Weaning and discontinuation of oxygen therapy for this deterioration (grade D).

U1. Lower the oxygen concentration if the patient is U9. Some patients may have episodic hypoxaemia (eg, copyright. clinically stable and the oxygen saturation is above after minor exertion or due to mucus plugging) the target range or if it has been in the upper zone after they have safely discontinued oxygen therapy. of the target range for some time (usually 4– An on-going prescription for a target saturation 8 hours) (grade D). range will allow these patients to receive oxygen as U2. If the target saturation is maintained, the new deliv- the need arises but transient asymptomatic desatur- fl

ery system and ow should be continued. Repeat ation does not require correction (grade D). http://bmjopenrespres.bmj.com/ blood gas measurements are not required. If the V. Practical aspects of oxygen use in prehospital and hospital patient is stable, the process can be repeated and care and use of oxygen alert cards (see full Guideline sections the patient can eventually be weaned off oxygen 9–11) (see section 12) (grade D). V1. Emergency oxygen should be available in primary U3. Most stable convalescent patients will eventually be care medical centres, preferably using oxygen cylin- stepped down to 2 L/min via nasal cannulae prior ders with integral high-flow regulators. Alternatively, to cessation of oxygen therapy. Patients at risk of oxygen cylinders fitted with high-flow regulators hypercapnic respiratory failure may be stepped (delivering up to 15 L/min) must be used to allow down to 1 L/min (or occasionally 0.5 L/min) via use with reservoir masks (grade D). nasal cannulae or a 24% Venturi mask at 2 L/min V2. Healthcare organisations should take measures to as the lowest oxygen concentration prior to cessa- eliminate the risk of oxygen tubing being con- on September 24, 2021 by guest. Protected tion of oxygen therapy (grade D). nected to the incorrect wall oxygen outlet or to U4. Oxygen therapy should be stopped once a patient outlets that deliver compressed air or other gases is clinically stable on low-concentration oxygen, and instead of oxygen. Air flow metres should be the oxygen saturation is within the desired range removed from the wall sockets or covered with a on two consecutive observations (but the prescrip- designated air outlet cover when not in use. Special tion for a target saturation range should remain care should be taken if twin oxygen outlets are in active in case of future deterioration). It may be use (grade D). appropriate to alter the target range following Good practice points related to practical aspects of oxygen senior review in patients with chronic cardiopul- therapy monary disease who either have saturations <94% Assessment and immediate oxygen therapy: when stable or in whom it is deemed sensible to ✓ Chronically hypoxaemic patients with a clinical discharge from hospital with saturations <94% exacerbation associated with a 3% or greater fall in pending an outpatient oxygen assessment. Oxygen oxygen saturation on their usual oxygen therapy

O’Driscoll BR, Howard LS, Earis J, et al. BMJ Open Resp Res 2017;4:e000170. doi:10.1136/bmjresp-2016-000170 17 BMJ Open Resp Res: first published as 10.1136/bmjresp-2016-000170 on 15 May 2017. Downloaded from Open Access

should usually be assessed in hospital with blood gas respiratory failure and carries an oxygen alert card. estimations. PaO2 of <7 kPa equates to SpO2 below Use of oxygen in these patients will be guided by the ∼85%. instructions on the alert card or by a Patient Specific ✓ The initial oxygen therapy to be used in the various Protocol which can be shared by hospital teams, the clinical situations is given in tables 1–4. ambulance service and the primary care team. ✓ If there is a clear history of asthma or heart failure or W. Practical aspects of oxygen dispensing, documentation and other treatable illness, appropriate treatment should monitoring be instituted in accordance with guidelines or stand- W1. Registered nurses and others who dispense drugs ard management plans for each disease. in hospitals should sign the drug chart or elec- ✓ The oxygen saturation should be monitored continu- tronic prescribing record at every drug round and ously until the patient is stable or arrives at hospital check that the patient is receiving oxygen therapy. for a full assessment. The oxygen concentration This is to check that the patient is within the target should be adjusted upwards or downwards to main- saturation and also to check whether weaning and tain the target saturation range. discontinuation should be instituted (grade D). ✓ In most emergency situations, oxygen is given to W2. Most patients are prescribed an oxygen target patients immediately without a formal prescription range. If patients are on air at the time of the drug or drug order. The lack of a prescription should round, registered nurses should sign the drug chart never preclude oxygen being given when needed in using a code such as ‘A’ for air and the observation an emergency situation. However, a subsequent chart should also be filled in using the code A for writtenrecordmustbemadeofwhatoxygen air (see table 5 and figure 19 in the full guideline) therapy has been given to every patient (in a similar (grade D). manner to the recording of all other emergency W3. All patients should have their oxygen saturation treatment). observed for at least 5 min after starting oxygen ✓ General practitioners or first responders visiting a therapy or for patients who require an increased patient’s home should carry a portable pulse oxim- concentration of oxygen and after oxygen therapy eter to assess hypoxaemia and guide use of oxygen if has been decreased or stopped (grade D). available and should call emergency services if hypox- W4. If the oxygen saturation is above the target saturation

aemia or other serious illness is suspected. range and the patient is stable, the delivery system or copyright. ✓ Those attending patients as an emergency in rural or oxygen flow rate should be modified to return the remote areas should consider carrying a portable saturation to within the target range (grade D). oxygen cylinder as part of their emergency equipment. W5. Patients who have a target saturation of 88–92% Oxygen alert cards for patients with hypercapnic should have their blood gases measured within respiratory failure: 30–60 min. This is to ensure that the carbon ✓

Patients with COPD (and other at-risk conditions) dioxide level is not rising. This recommendation http://bmjopenrespres.bmj.com/ who have had an episode of hypercapnic respiratory also applies to those who are at risk of developing failure should be issued with an oxygen alert card hypercapnic respiratory failure but who have a and with a 24% or 28% Venturi mask. They should normal PCO2 on the initial blood gas measurement be instructed to show the card to the ambulance (grade D). crew and emergency department staff in the event of W6. Stable patients whose oxygen saturation is within an exacerbation. their target saturation range of 94–98% do not ✓ Oxygen alert cards with agreed content can be need repeat blood gas measurements within obtained via the BTS website. 30–60 min if there is no risk of hypercapnic respira- ✓ The content of the alert card should be specified by tory failure and acidosis and may not need any the physician in charge of the patient’s care, based further blood gas measurements unless there should on previous blood gas results. be further deterioration including symptoms or on September 24, 2021 by guest. Protected ✓ The primary care team and ambulance service signs of possible hypercapnia (grade D). should also be informed by the hospital COPD team W7. Stable patients on oxygen treatment should have that the patient has had an episode of hypercapnic SpO2 and physiological variables (eg, NEWS) mea- respiratory failure and carries an oxygen alert card. sured four times a day (grade D). The home address and ideal oxygen concentration W8. In those who have signs of critical illness (eg, or target saturation ranges of these patients can be NEWS score 7 or above), oxygen saturation should flagged in the ambulance control systems and infor- be monitored continuously and the patient may mation disseminated to ambulance crews when require level 2 or level 3 care on a high- required. dependency unit or critical care unit (grade D). ✓ When possible, out-of-hours services providing emer- W9. If the patient is clinically stable and the oxygen sat- gency primary care services should be informed by uration is within the target range, treatment should the hospital COPD team or by the primary care team be continued (or eventually lowered) depending that the patient has had an episode of hypercapnic on the clinical situation (grade D).

18 O’Driscoll BR, Howard LS, Earis J, et al. BMJ Open Resp Res 2017;4:e000170. doi:10.1136/bmjresp-2016-000170 BMJ Open Resp Res: first published as 10.1136/bmjresp-2016-000170 on 15 May 2017. Downloaded from Open Access

W10. Oxygen therapy should be increased if the saturation include basic training in oxygen use in the mandatory is below the desired range and decreased if the sat- training programmes for all clinical staff (grade D). uration is above the desired range (and eventually discontinued as the patient recovers) (grade D). Healthcare providers need to use clinical judge- W11. The new saturation (and the new delivery ment, knowledge and expertise when deciding system) and flow rate should be recorded on the whether it is appropriate to apply recommendations patient’s observation chart after 5 min of treat- for the management of patients. The recommenda- ment at the new oxygen concentration. Each tions cited here are a guide and may not be appro- change should be recorded by the clinician priate for use in all situations. The guidance trained to administer oxygen by signing the provided does not override the responsibility of observation chart (only changes should be healthcare professionals to make decisions appropri- signed for) (grade D). ate to the circumstances of each patient, in consult- W12. Repeat blood gas measurements are not required ation with the patient and/or their guardian or carer. for stable patients who require a reduced concentra- tion of oxygen (or cessation of oxygen therapy) to Collaborators BTS Emergency Oxygen Guideline Group: BROD, LH, JE, VM, maintain the desired target saturation (grade D). Dr Sabrina Bajwah, Professor Richard Beasley, Dr Katrina Curtis, Professor W13. Patients with no risk of hypercapnic respiratory Anthony Davison, Dr Alistair Dorward, Dr Chris Dyer, Angela Evans, Lucy failure do not always need repeat blood gas mea- Falconer, Clare Fitzpatrick, Dr Simon Gibbs, Dr Kim Hinshaw, Dr Robin surements after an increase in oxygen concentra- Howard, Dr Binita Kane, Dr Jeff Keep, Carol Kelly, Hasanin Khachi, Dr tion. However, the patient requires clinical review Muhammed Asad Iqbal Khan, Dr Roop Kishen, Leigh Mansfield, Dr Bruce Martin, Dr Fionna Moore, Dr Duncan Powrie, Dr Louise Restrick, Professor to determine why the oxygen saturation has fallen Christine Roffe, Dr Mervyn Singer, Dr Jasmeet Soar, Dr Iain Small, Lisa Ward, (grade D). David Whitmore, Professor Wisia Wedzicha, Dr Meme Wijesinghe. W14. Patients at risk of hypercapnic respiratory failure – Contributors BROD was lead author responsible for the overall document and (usually those with a target range of 88 92%; see production of supporting appendices. LH was lead author for sections 3–6. JE table 4) require repeat blood gas assessment 30– was lead author for section 14. VM was lead author for sections 9–11. All 60 min after an increase in oxygen therapy (to authors were responsible for the final approval of the guideline. Members of the ensure that the carbon dioxide level is not rising) guideline group and their contributions to individual sections of the guideline

(grade D). are listed in Annex 1 of the full guideline. copyright. W15. For patients with no risk of hypercapnic respira- Competing interests LH has received research funding support from Bayer tory failure, monitoring by pulse oximeter is suffi- PLC. VM has received lecture and meeting attendance support from Chiesi, GSK and AstraZeneca. cient (repeated blood gases not required) provided the patient is clinically stable and the Provenance and peer review Not commissioned; internally peer reviewed. oxygen saturation remains in the desired range, Data sharing statement No additional data are available. –

usually 94 98% (grade D). http://bmjopenrespres.bmj.com/ W16. If a patient’s oxygen saturation is lower than the prescribed target range, first check all aspects of ONLINE APPENDICES—AVAILABLE AT http://www.brit- the oxygen delivery system and the oximeter thoracic.org.uk device for faults or errors (grade D). W17. If a patient’s oxygen saturation is consistently 1. Search strategy lower than the prescribed target range, there 2. Evidence tables 3. Summary of guideline for hospital use should be a medical review and the oxygen 4. Example of local oxygen policy therapy should be increased according to an 5. Summary of guideline and flow charts for emergency oxygen use agreed written protocol (grade D). in ambulances, community and prehospital settings W18. If the oxygen saturation fails to rise following 6. Patient information sheet 5–10 min of increased oxygen therapy or if there is 7. Slides on emergency oxygen use for doctors on September 24, 2021 by guest. Protected 8. Slides on emergency oxygen use for nurses, midwives, pharma- clinical concern following medical review, then blood cists, physiotherapists and other practitioners who use oxygen gasmeasurementsshouldberepeated(gradeD). 9. Key points for hospital managers and oxygen champions X. Training in oxygen prescribing and use 10. Key points for primary care managers, clinical commissioning groups X1. All clinicians prescribing oxygen should have 11. Dissemination and implementation of the guideline appropriate training and access to written or elec- tronic oxygen prescribing guidelines based on this REFERENCES national guideline (grade D) (see online 1. O’Driscoll BR, Howard LS, et al. BTS Guideline for oxygen use in appendices 7 and 8). adults in healthcare and emergency settings, Thorax 2017;72 X2. Every hospital should have a training programme to (Suppl 1):pi1–i89. 2. O’Driscoll BR, Howard LS, Davison AG, et al. BTS guideline for ensure that clinical staff are familiar with the hospital’s emergency oxygen use in adult patients. Thorax 2008;63(Suppl 6): oxygen administration policies. In view of the high vi1–68. 3. Hardinge M, Annandale J, Bourne S, et al. British Thoracic Society number of adverse incidents related to oxygen guidelines for home oxygen use in adults. Thorax 2015;70(Suppl 1): therapy, it is recommended that all acute trusts should i1–43.

O’Driscoll BR, Howard LS, Earis J, et al. BMJ Open Resp Res 2017;4:e000170. doi:10.1136/bmjresp-2016-000170 19 BMJ Open Resp Res: first published as 10.1136/bmjresp-2016-000170 on 15 May 2017. Downloaded from Open Access

4. Hale KE, Gavin C, O’Driscoll BR. Audit of oxygen use in emergency 10. Boyle M, Wong J. Prescribing oxygen therapy. An audit of ambulances and in a hospital emergency department. Emerg Med J oxygen prescribing practices on medical wards at North 2008;25:773–6. Shore Hospital, Auckland, New Zealand. NZMedJ2006;119: 5. Wilson AT, Channer KS. Hypoxaemia and supplemental oxygen U2080. therapy in the first 24 hours after myocardial infarction: the role of 11. Kbar FA, Campbell IA. Oxygen therapy in hospitalized patients: the pulse oximetry. J R Coll Physicians Lond 1997;31:657–61. impact of local guidelines. J Eval Clin Pract 2006;12:31–6. 6. Cook DJ, Reeve BK, Griffith LE, et al. Multidisciplinary education for 12. AGREE II Instrument. Secondary AGREE II Instrument. http://www. oxygen prescription. A continuous quality improvement study. Arch agreetrust.org/agree-ii/. Intern Med 1996;156:1797–801. 13. JRCALC. UK Ambulance Services Clinical Practice Guidelines 7. Small D, Duha A, Wieskopf B, et al. Uses and misuses of oxygen in 2013. London: JRCALC, 2013. hospitalized patients. Am J Med 1992;92:591–5. 14. BTS. British Thoracic Society Guideline Production Manual. London: 8. Dodd ME, Kellet F, Davis A, et al. Audit of oxygen prescribing before BTS, 2014:31. and after the introduction of a prescription chart. BMJ 15. COPD NICE Clinical Guideline 101, 2010. 2000;321:864–5. 16. Davidson AC, et al. BTS/ICS guideline for the ventilatory management 9. Howell M. An audit of oxygen prescribing in acute general medical of acute hypercapnic respiratory failure in adults. Thorax 2016;(Suppl 2) wards. Prof Nurse 2001;17:221–4. 71ii1–ii35. copyright. http://bmjopenrespres.bmj.com/ on September 24, 2021 by guest. Protected

20 O’Driscoll BR, Howard LS, Earis J, et al. BMJ Open Resp Res 2017;4:e000170. doi:10.1136/bmjresp-2016-000170