P10. Oxygen therapy

Long-term oxygen therapy has survival benefits for COPD patients with hypoxaemia [evidence level I, strong recommendation].

Long term oxygen therapy (LTOT) reduces mortality in COPD (Medical Research Council Working Party 1981Nocturnal Oxygen Therapy Trial Group 1980, Gorecka 1997, Zielinski 1998, American Thoracic Society 1995, Siafakas 1995, Tarpy 1995). It may also have a beneficial impact on haemodynamics, haematological status, exercise capac­ity, lung mechanics and mental state (Weitzenblum 1985, Zielinski 1998Tarpy 1995). Although effective, it is a potentially expensive and cumbersome therapy that should only be prescribed for those in whom there is evidence of benefit (see below). Information on prescribing oxygen therapy is given in Appendix 3.

Long-term continuous oxygen therapy: (ideally at least 18 hours a day) is appropriate for patients who have PaO2 consistently < 55 mmHg (7.3 kPa; SpO2 less than 88%) (Medical Research Council Working Party 1981Nocturnal Oxygen Therapy Trial Group 1980) when breathing air, at rest and awake [evidence level I]. If oxygen is prescribed when the patient’s condition is unstable (e.g., during an exacerbation), then the requirement for it should be reviewed four to eight weeks after initiation as it has been demonstrated in several studies that patients frequently do not fulfil the criteria for LTOT at subsequent follow up (Eaton 2004, Levin 2018).  In the latter study 54% of patients no longer required LTOT at review 1-2 months after discharge. At assess­ment for ongoing therapy, the patient’s condition must be stable, all potentially reversible factors must have been treated and the patient must have stopped smoking at least one month previously.

Polycythaemia (haemoglobin level > 170 g/L), clinical or electrocardiographic evidence of pulmonary hypertension, as well as episodes of right heart failure, are consistent with the systemic effects of chronic hypoxaemia, and continuous oxygen should be supplied if the stable PaO2 is 55– 59 mmHg (7.3–7.9 kPa; SpO2 < 90%) (Siafakas 1995American Thoracic Society 1995). Continuous oxygen therapy is of most benefit for patients with increased arterial PaCO2 (> 45 mmHg, or 6 kPa) (Nocturnal Oxygen Therapy Trial Group 1980).

Government funding is available on the basis that the prescribing doctor is an approved prescriber (usually a respiratory physician). Oxygen is usually supplied to patients meeting specific criteria and means testing by state or regional health departments in Australia and New Zea­land (Serginson 2009).

Oxygen in patients with moderate hypoxaemia

A large study of patients with moderate hypoxaemia (SpO2 89 to 93%) was powered originally to determine whether continuous oxygen therapy improved mortality (Long-Term Oxygen Treatment Trial Research 2016). Subsequently, inclusion criteria were altered to include those who desaturated with exertion but were minimally hypoxaemic at rest (SpO2 ≥ 94% resting but desaturating to <90% for >10 seconds and with SpO2 ≥ 80% for ≥ 5 mins).  The study demonstrated no difference between groups in the composite outcome of mortality or time to first hospitalisation, nor in any other outcome including quality of life.

738 participants were randomised to receive oxygen at 2 litres per minute or no oxygen. 57% had resting hypoxaemia and were prescribed continuous oxygen at 2 litres per minute and 43% were prescribed oxygen at 2 litres per minute during exercise and sleep. Over a median follow-up of 18.4 months, the median use of oxygen was 15.1 ± 6.2 hours per day in the continuous group and 11.3 ± 5 hours per day in the exercise and nocturnal group. 51 adverse events were noted, with three patients requiring hospitalisation on account of these. The majority of adverse effects were slips and falls, but fire and burns also occurred.

Limitations to this study included an absence of blinding, no placebo arm, and lack of clarity as to whether the study was adequately powered for the modified composite primary outcome.

The findings from this study and its accompanying editorial are consistent with clinical practice guidelines on adult domiciliary oxygen provided by the Thoracic Society of Australia and New Zealand which recommend provision of long-term continuous oxygen therapy only in those who are significantly hypoxaemic (see P10 above) and recommend use of ambulatory oxygen only in the few patients who demonstrate benefit in a blinded test (McDonald 2016).

In a systematic review and meta-analysis of the effects of home oxygen (delivered either as LTOT or nocturnally) in patients with moderate hypoxaemia, six high-quality studies were included. The results demonstrated the effect of home oxygen in reducing 3-year mortality was small or absent and the authors concluded the data did not support the widespread use of home oxygen in this population of patients with moderate hypoxaemia (Lacasse 2022) [evidence level I].

Ambulatory oxygen therapy

In patients who qualify for long-term oxygen therapy (LTOT), ambulatory oxygen therapy can be used in order to maximize usage achieve an average usage of 18 hours day (Nocturnal Oxygen Therapy Trial Group 1980.

In patients who do NOT quality for LTOT, available evidence does not allow any firm conclusions to be made about the use of long-term intermittent ambulatory domiciliary oxygen therapy in patients with COPD who do not meet the criteria for LTOT. This conclusion is based on a Cochrane Review comprising four studies (total of 331 patients) (Ameer 2014) who received oxygen or air (blinded) for between two and 12 weeks in the home setting. This review found no significant difference in exercise tolerance or mortality in those receiving supplemental oxygen compared to breathing air supplied by a cylinder. Although statistically significant benefits favouring oxygen were found in Health-related Quality of Life (HRQoL) (dyspnoea and fatigue domains of the Chronic Respiratory Disease Questionnaire (CRQ)) the improvements did not reach the threshold for clinical significance. A clinically significant reduction in end exercise dyspnoea favouring oxygen was found in two studies [evidence level I].

Ambulatory oxygen should not be routinely offered to patients who are not eligible for LTOT. However, the use of short-term intermittent oxygen therapy may be considered for:

People who experience oxygen desaturation on exertion

A Cochrane Review of 31 studies found that ambulatory oxygen was efficacious in single assessment studies (in the hospital or laboratory setting) when comparing an exercise test performed breathing oxygen or air in patients with moderate to severe COPD (Bradley 2005) [evidence level I].  Benefits were shown in endurance exercise capacity, dyspnoea at isotime and oxygen saturation. However, the minimum clinically important difference in these variables with oxygen therapy is unknown.  Due to the heterogeneity of the studies, subgroup analyses were not possible to determine which patients were more likely to benefit.  Acute benefit may be established by comparing exercise tolerance, oxygen saturation and dyspnoea on a field walk test or treadmill test when breathing oxygen and when breathing air (blinded). A cycle ergometry test should not generally be used for this purpose as oxygen desaturation is significantly greater in COPD patients when walking as compared to cycling (Turner 2004, Poulain 2003). It is important to consider that most patients will walk further on a repeat walk test and hence a practice test is usually necessary (Singh 2014c). The endurance shuttle walk test (ESWT) has been shown to be more responsive than the 6-minute walk test when assessing the benefits of ambulatory oxygen (Revill 2010) and it would appear that a practice ESWT may not be necessary when two ESWTs are performed on the same day (Singh 2014c).  However, the ESWT requires patients to first perform the incremental shuttle walk test in order to determine the walking speed for the ESWT. Ideally, the oxygen system used in the assessment should be the same as the system the patient would use if oxygen were prescribed at home (e.g. trolley or shoulder bag to transport the cylinder). It is to be noted that short-burst oxygen i.e. oxygen inhaled immediately prior and/or following exertion with the aim of relieving breathlessness or improving exercise tolerance is not effective (O’Neill 2006, O’Driscoll 2008) [evidence level II].

The prescription of supplemental oxygen should not be based solely on an improvement in the distance achieved on a walk test. Factors such as a reduction in dyspnoea and agreement to use oxygen within the home and outdoors during activity should also be considered. As the relationship between single assessments and long-term benefits is unclear, the acute assessment should form only part of the determination and benefit of ongoing ambulatory oxygen therapy. Long-term review and determination of oxygen usage are also important (Bradley 2007).

Ambulatory oxygen therapy during pulmonary rehabilitation

In the absence of need for LTOT there is no direct evidence that the treatment of exercise-induced hypoxaemia retards long-term pulmonary hypertension or prolongs life. Nevertheless, in patients who desaturate during exercise training, supplemental oxygen has been proposed with the aim of delaying the onset of dynamic hyperinflation and the associated dyspnoea (O’Donnell 1997, O’Donnell 2001), and to improve the benefit from training (Emtner 2003).  However, neither a 2019 RCT (Alison 2019), nor a systematic review of earlier studies (Nonoyama 2007) [evidence level I] support this approach.  In the Australian RCT (Alison 2019), 111 subjects with moderate to severe COPD who had oxygen desaturation to <90% during 6-minute walk tests were randomised to either air or oxygen via nasal prongs at 5L/minute for 8 weeks of 3X/week treadmill and cycle exercise sessions.  Both groups improved with respect to outcomes of Chronic Respiratory Disease Questionnaire and endurance shuttle walk test, however there was no additional benefit with supplementary oxygen. This RCT provides strong evidence that the provision  of supplementary oxygen does not improve these important outcomes in such exercise programs even when subjects are known to desaturate to <90%.

Other indications for intermittent oxygen therapy

Patients living in isolated areas or prone to sudden life-threatening episodes while they are awaiting medical attention or evacuation by ambulance;

Patients travelling by air. Flying is generally safe for patients with chronic respiratory failure who are on long- term oxygen therapy, but the flow rate should be increased by 1-2 L/minute during the flight (see also below).

Nocturnal oxygen therapy: A large multicentre randomised controlled trial of nocturnal oxygen therapy versus air delivered via concentrator or sham concentrator (the so-called INOX trial) was performed in patients with COPD who did not fulfil criteria for LTOT (Lacasse 2020). Exclusion criteria included smoking cessation less than 6 months previously, significant obstructive sleep apnoea (AHI>15), BMI>40, known left heart failure and bronchiectasis. Inclusion criteria included desaturating to SPO2<90% for at least 30% of the recording time on nocturnal oximetry. Recruitment to this trial was stopped early because of recruitment and retention difficulties after n=243 patients, of a planned n=600, had undergone randomisation. At three years of follow up there was no difference between the groups  in the composite endpoint of death from any cause or a requirement for long-term oxygen therapy as defined by the Nocturnal Oxygen Therapy Trial (NOTT) criteria in the intention-to-treat population.

Although this trial was underpowered, based on these results and those of two previous studies by Fletcher et al (Fletcher 1992) and Chaouat et al (Chaouat 1999) current evidence does not support the prescription of nocturnal oxygen therapy to improve survival or slow disease progression in patients with COPD. However, the confidence intervals around the pooled treatment effects from a meta-analysis of these three studies performed by the authors of this recent INOX trial and presented as supplementary to this study concluded that  the confidence limits around these outcomes are wide, and clinically significant effects are plausible [evidence level I]. More research is needed. In the meantime, some patients with hypoxaemia during sleep may benefit  from nocturnal oxygen therapy. Nocturnal hypoxaemia should be considered in patients whose arterial gas tensions are satisfactory when awake, but who have daytime somnolence, polycythaemia or right heart failure. Oxygen may be indicated for patients whose nocturnal arterial oxygen saturation repeatedly falls below 88%. Sleep apnoea should be excluded and treated independently.