The range of PaO2 between 7.3 and 8 kPa (55 mmHg and 60 mmHg) remains a grey area. From the MRC and NOT Trial, it is not clear whether LTOT has a beneficial effect on prognosis for these patients. This question was addressed by a randomized controlled trial in 135 Polish patients with COPD . These patients had a PaO2 between 7.4kPa (55.5 mmHg) and 8.7kPa (65.3 mmHg), with mean PaO2 8 kPa (60 mmHg). LTOT, over at least 3 years, was not associated with a survival benefit, and there was no difference in survival between the patients with a PaO2 above or below 8 kPa (60 mmHg). However, the mean daily use of oxygen was only 13.5h.
Patients with a PaO2 between 7.3 kPa and 8 kPa (55 mmHg and 60 mmHg) and secondary polycythaemia, peripheral oedema, evidence of pulmonary hypertension or nocturnal hypoxaemia (defined as oxygen saturation (SaO2) below 90% for at least 30% of the night ), are specifically recommended for LTOT .
Secondary polycythaemia and peripheral oedema are relatively easy to detect clinically and were features of the patients in the MRC study and the NOT Trial [1,2]. This is not the case for either pulmonary hypertension or nocturnal hypoxaemia. Patients with pulmonary hypertension were included in the NOT Trial, but not in the MRC study. Pulmonary hypertension was defined electrocardiographically by a 3-mm P pulmonale in leads II, III and aVF . The diagnosis of pulmonary hypertension clinically is imprecise and is left at the discretion of the clinician in the 1999 UK guidelines . The majority of patients with severe COPD and a PaO2 of less than 8 kPa (60 mmHg) will have some pointers of pulmonary hypertension — for example, prominent pulmonary arteries on their chest radiograph. Those clinicians wanting to prescribe LTOT for patients with a PaO2 up to 8 kPa (60 mmHg) can therefore do so within the guidelines, by choosing a lower threshold for making a clinical diagnosis of pulmonary hypertension. However, there is no direct evidence for
LTOT being of benefit in this group of patients with a clinical diagnosis of pulmonary hypertension.
The specific criterion of nocturnal hypoxaemia with a PaO2 between 7.3 kPa and 8kPa (55mmHg and 60mmHg) for the provision of LTOT is more controversial. Nocturnal hypoxaemia due to nocturnal hypoventilation is well recognized in COPD and can lead to pulmonary hypertension [6,7]. There are two randomized controlled trials of nocturnal oxygen in patients with COPD and nocturnal hypoxaemia [8,9]. However, in both studies, the mean PaO2 at entry was above 8kPa (60mmHg). In the Fletcher et al. study of seven patients treated with nocturnal oxygen and nine patients given sham oxygen, with a mean PaO2 at entry of about 10kPa (75 mmHg), there was no effect of oxygen on mortality over 3 years . There were no significant differences in pulmonary artery pressure between the nocturnal oxygen and control groups, either at the start or end of the study. However, at 3 years there was a small increase in pulmonary artery pressure in the untreated group and a small reduction in pulmonary artery pressure in the group treated with nocturnal oxygen; the change in pulmonary artery pressure was significantly different between the groups.
In the second larger study of 76 patients with nocturnal desaturation, with a mean PaO2 at entry of 8.4kPa (63mmHg), there was also no difference in mortality at 2years between patients receiving nocturnal oxygen and those who did not . Nocturnal oxygen did not allow delay in the prescription of LTOT and had no effect on pulmonary haemodynamics in this study. There are no controlled trials of patients with nocturnal hypoxaemia and a daytime PaO2 between 7.3 kPa and 8kPa (55 mmHg and 60 mmHg). While LTOT may prevent, delay, or reverse the progression of pulmonary hypertension , this may not be of large enough benefit to translate into improved survival. The main determinant of survival in COPD is the severity of airway obstruction, and the only measure that has an effect on progression of airway obstruction is smoking cessation.
The recommendation to prescribe LTOT for patients with moderate hy-poxaemia — PaO2 between 7.3 kPa and 8 kPa (55 mmHg and 60 mmHg) — and clinically determined pulmonary hypertension or nocturnal hypoxaemia is therefore based on extrapolation of physiological data, rather than data showing improved survival. This recommendation has service and cost implications that need to be taken into account.
The implication of including nocturnal hypoxaemia as a criterion for prescribing LTOT in COPD is that patients with a daytime PaO2 between 7.3 kPa and 8 kPa (55 mmHg and 60 mmHg) will have to be screened for nocturnal hypoxaemia with overnight oximetry. The prevalence of nocturnal hypoxaemia in this group is high, and its severity is directly related to the severity of the daytime hypoxaemia . Nocturnal hypoxaemia was present in 43% of patients with a daytime PaO2 between 8kPa and 9.3kPa (60mmHg and 69.8 mmHg) . The prevalence is likely to be even higher in those with a daytime PaO2 between 7.3kPa and 8kPa (55mmHg and 60mmHg). This will therefore require significant resources to implement.
A more cost-effective use of resources to improve prognosis in COPD is likely to be screening for severe hypoxaemia in patients with severe COPD in the community, so that all patients with a PaO2 less than 7.3 kPa (55 mmHg) receive LTOT. It has been estimated that there may be 60 000 individuals with COPD in England and Wales who would meet the criteria for the prescription of LTOT . However, the numbers of patients on LTOT are much less than the estimates . Currently, there are approximately 20000 patients in the UK with oxygen concentrators. Screening by oximetry in the community has been shown to be effective in practice , and the benefit of LTOT for this group of patients is proven. The Cochrane review  of LTOT for patients with COPD and a PaO2 below 7.3 kPa (55 mmHg) calculated a number needed to treat (NNT) of 4.5 from the MRC study , i.e. treating five patients with LTOT would save one life over 5 years. Identifying and treating those patients with COPD who are not on LTOT and should be is a serious problem that can and needs to be addressed as a high priority.
Patients with COPD and additional reasons for nocturnal hypoxaemia — e.g. previous thoracoplasty, obstructive sleep apnoea—need to have formal sleep studies performed, as oximetry is not sufficient to evaluate these patients . Nocturnal oxygen therapy may be prescribed with non-invasive ventilation (NIV) or continuous positive airway pressure (CPAP) for some of these patients.
Was this article helpful?