Oxygen therapy in chronic lung diseaseBMJ 1998; 317 doi: https://doi.org/10.1136/bmj.317.7162.871 (Published 26 September 1998) Cite this as: BMJ 1998;317:871
- P J Rees,
- F Dudley
Short term oxygen therapy
Oxygen in first aid
Oxygen can be used in acute situations where there is hypoxia because of acute lung injury or reduced ventilation, when there is underlying chronic lung disease, or when it is important to maintain oxygen delivery to certain tissues. Maintaining oxygen delivery may help tissue survival when there is cardiac failure with pulmonary oedema, ischaemic tissue (such as after an infarction), or a cerebrovascular accident. Oxygen saturation should be monitored by pulse oximetry. If oxygen saturation is low additional inspired oxygen should be used to maintain saturation near 100%.
Oxygen can be used empirically in acute situations if a patient is cyanosed or has a raised respiratory rate. Generally it is safe to use a high inspired oxygen concentration as long as there is no suggestion that the patient has chronic obstructive pulmonary disease. The main danger of acute oxygen is that carbon dioxide retention may occur in some patients with chronic obstructive pulmonary disease. If this is suspected the oxygen should be given by a high flow 24% or 28% mask until arterial blood gases can be measured. Oxygen increases the fire hazard at the scene of an accident.
In emergency situations lack of availability of oxygen and an appropriate delivery system may prevent early treatment. Ambulances carry oxygen and paramedical staff are trained to use the equipment in acute situations. Some general practitioners also carry small cylinders that can deliver 2 l/min for two hours, which avoids the need to wait for an ambulance.
Occasional oxygen in chronic lung disease
Breathlessness in chronic lung conditions or in carcinoma of the bronchus may be intermittent. It is likely to increase on exertion, and simple everyday tasks such as washing, dressing, or eating may cause problems. Some patients feel that a short period on oxygen before or just after such exertions reduces their breathlessness. It is more appropriate, however, to use the oxygen while the oxygen demand is increased. For most of these tasks, especially eating, nasal cannulae are a more convenient form of delivery than face masks.
In severely ill patients occasional use of oxygen can be based on patients' reported need and followed up with blood gas analysis to ensure there is no serious carbon dioxide retention
Use of intermittent oxygen can be made more rational by measuring baseline saturation or arterial oxygen levels and assessing the effect of supplementation on the measurements and on the ability to perform the task. A purer approach is to compare the effects of supplemental oxygen and air in a double-blind fashion using patients as their own control.
Relief of exercise related breathlessness
The addition of oxygen during exercise reduces ventilation at a given workload in normal subjects and in those with chronic respiratory disease. This is achieved by reducing the respiratory rate, which relieves breathlessness and increases exercise tolerance. Studies on exercise in chronic obstructive pulmonary disease show an increase in endurance of 30-50% with oxygen. Any extra exertion involved in carrying equipment such as cylinders will reduce the benefit. Lightweight liquid oxygen systems are the most useful; heavier cylinders can be wheeled on a trolley.
Use of oxygen can increase the ability to do everyday tasks in chronic lung disease, but it is often limited by availability of suitable portable systems
Most suitable patients have a PaO2 at rest <8.5 kPa or a drop in saturation to less than 90% on exercise. The benefit can be assessed in a walking test. This can be done as a maximum distance in a given time (conventionally 6 or 12 minutes), a maximum walk before stopping, or a timed shuttle test. Performance improves with familiarity over the first two or three such tests, and a good positive effect would be the ability to reach an important everyday goal or a 50% increase in exercise. Providing oxygen during exercise is difficult and should be assessed carefully. Usually this assessment is carried out in a respiratory function laboratory.
Oxygen used in this way improves quality of life and the ability to perform everyday tasks rather than increasing survival, which needs prolonged use of oxygen every day.
Dangers of carbon dioxide retention
In chronic lung disease it is important to consider the possibility of carbon dioxide retention when breathing an increased fraction of oxygen. One reason for a rise in carbon dioxide is that hypercapnic respiratory drive has been blunted by chronic carbon dioxide retention. Hypoxic drive becomes more important so that an increase in inspired oxygen improves hypoxia, reducing ventilation and increasing carbon dioxide pressure. In addition, increased inspired oxygen may overcome local pulmonary vasoconstriction in the lung increasing wasted perfusion of poorly ventilated lung with greater mismatch of ventilation and perfusion. When chronic oxygen use is proposed the patient's arterial blood gases must be measured on and off the planned oxygen regimen.
In chronic lung disease problems with carbon dioxide retention are unlikely if the resting carbon dioxide pressure is normal
Long term oxygen therapy
Chronic hypoxia produces a mixture of permanent and reversible structural changes in the pulmonary vasculature. There is hypertrophy of the muscular media in small pulmonary arteries, muscularisation of pulmonary arterioles, and fibrosis of the intimal layer. In stable disease initiation of oxygen therapy causes some pulmonary artery vasodilatation in areas of the lung with poor ventilation; this causes a small increase in the ratio of dead space to tidal volume. Minute ventilation at rest is unchanged but is reduced on exercise.
Reversal of the profound hypoxia of severe chronic obstructive pulmonary disease has been found to reduce mortality if the oxygen is used for a minimum of 15 hours each day. Correction of hypoxia may have other benefits such as reducing polycythaemia and reducing or preventing progression of pulmonary hypertension together with an increase in exercise tolerance.
What the trials show
Two thirds of the patients on long term oxygen therapy in the United Kingdom have chronic obstructive pulmonary disease. Most of the evidence for this treatment comes from two major trials published at the start of the 1980s. In the Medical Research Council study published in 1981 there were 87 patients with a mean age of 58 years and mean FEV1 of 0.7. They had modest pulmonary hypertension with a mean pressure of 34 mm Hg, PaO2 of 6.8 kPa, and PaCO2 of 7.3 kPa. The active group took oxygen 15 hours daily compared with none in the control group. Thirty of the 45 in the control group died compared with 19 out of 42 in the oxygen group.
UK criteria for long term oxygen
PaO2 <7.3 kPa
PaCO2 >6 kPa
FEV1 <1.5 l
FVC <2.0 l
Measurements should be stable over three weeks and taken with patient receiving optimal medical treatment
The predicted five year survival in the control group was only 18% and the average annual percentage mortality risk was 30% reflecting the poor prognosis of this condition. The pulmonary artery pressure was unchanged in the group on oxygen but rose by a mean of 2.7 mm Hg/year in the control group.
The nocturnal oxygen therapy trial conducted in North America recruited 208 patients with a mean age of 66 years, FEV1 of 0.7, and pulmonary artery pressure of 30 mm Hg. The study compared 12 hours' overnight oxygen with continuous oxygen (in practice 17.7 hours a day). The mean arterial gas pressures at entry were PaO2 6.8 kPa and PaCO2 5.8 kPa. Annual mortality was 21% in the overnight oxygen group and 11% in the continuous oxygen group. At six months the pulmonary artery pressure at rest in the continuous oxygen group had fallen slightly.
Implications of the studies
The two trials were not blinded by comparison with air cylinders but are accepted as adequate evidence that oxygen improves prognosis in chronic obstructive pulmonary disease. If oxygen is used for at least 15 hours daily the pulmonary artery pressure may fall and, at least, the expected rise seems to be prevented. The combination of the results of the two trials suggests that the nearer the use of oxygen is to 24 hours daily the better the outlook. In addition to the improved prognosis there are mild neuropsychological benefits and improvements in quality of life.
The mechanism of improvement in prognosis on oxygen remains unclear. Changes in pulmonary artery pressure are not great, and improved survival might be related to reduction in arrhythmias related to hypoxia. The effects of long term oxygen in chronic obstructive pulmonary disease have been extended to the treatment of other respiratory conditions. This is probably reasonable, although no real evidence exists.
When hypercapnia is prominent it may be appropriate to consider a combination of oxygen with non-invasive ventilatory support using a face mask and supplementary pressure during inspiration. Such treatment has proved useful in restrictive lung conditions, particularly those involving the chest wall and respiratory muscles.
Non-invasive ventilation has been used successfully in acute exacerbations of chronic obstructive pulmonary disease associated with hypercapnia and acidosis and also in long term home management. However, its precise role in long term management of chronic obstructive pulmonary disease has not been established.
The slight decrease in ventilation during sleep is associated with a slight drop in PaO2 and rise in PaCO2. When the PaO2 is initially low disturbances in ventilation, particularly those seen in rapid eye movement sleep, result in further periods of desaturation. These dips are typically scattered through the night at intervals of about two hours and are associated with transitory rises in pulmonary artery pressure. In patients on long term oxygen therapy the flow at night can be increased by 1 l/min to minimise these dips.
Oxygen has been shown to improve sleep quality and reduce electrocardiographic abnormalities, and in some countries nocturnal hypoxia alone is accepted as an indication for long term oxygen. Nocturnal hypoxia is found in a quarter of patients with chronic obstructive pulmonary disease with daytime PaO2 above 8 kPa, who would not meet most standard criteria for long term oxygen. It is associated with sustained higher pulmonary artery pressure and shorter survival. Nocturnal oxygen reduces the pulmonary artery pressure changes, but the use of overnight oxygen is not yet supported by the quality of survival evidence established for daytime hypoxia.
Patients with chronic obstructive pulmonary disease, who are predominantly male middle aged smokers and often overweight, may also have obstructive sleep apnoea. The low baseline saturation will mean greater drops in saturation with apnoea. Patients with both conditions are at particular risk of fluid retention and cor pulmonale. Patients with chronic obstructive pulmonary disease who have cor pulmonale despite a daytime PaO2 above 8 kPa should have oxygen saturation monitored continuously for a night to look for a sleep apnoea pattern or prolonged desaturation. This can be done at home with a suitable pulse oximeter.
F Dudley is a general practitioner, Hurley Clinic, London
The ABC of Oxygen is edited by Richard M Leach, consultant physician, and John Rees, consultant physician, Guy's and St Thomas's Hospital Trust, London.
Medical Research Council Working Party. Long term domiciliary oxygen therapy in chronic hypoxic cor pulmonale complicating chronic bronchitis and emphysema. Lancet 1981;i:681-6.
Nocturnal Oxygen Therapy Trial Group. Continuous or nocturnal oxygen therapy in hypoxic chronic obstructive lung disease. Ann Intern Med 1980;93:391-8.