LTOT reverses hypoxaemia and prevents hypoxia, and has been shown to improve life expectancy in patients with chronic lung disease.

A British Medical Research Council (MRC) study compared hypoxaemic patients receiving oxygen for 15 h•day-1, including the hours of sleep, with patients receiving no oxygen. This trial demonstrated that oxygen was associated with a significant reduction in mortality. The Nocturnal Oxygen Therapy Trial (NOTT), comparing continuous oxygen therapy (average 19 h•day-1) with therapy for 12 h•day-1, including the hours of sleep, demonstrated a further reduction in mortality using continuous oxygen.

The mechanism for improved survival has yet to be completely delineated, but pulmonary haemodynamics appear to play a role [3-7]. Oxygen therapy has been accompanied by a modest yearly decline in pulmonary artery pressure.

Pulmonary vascular resistance may be decreased in acute response to oxygen if the patient is nonoedematous, but not in patients with oedema.

Continuous oxygen therapy reverses secondary polycythemia, improves cardiac function during rest and exercise [8, 9], reduces the oxygen cost of ventilation, and improves exercise tolerance [10-12] and quality of life [13]. Hypoxemic patients (Pa,O2 50 mmHg) may experience neuropsychiatric deficits in abstract thinking, motor skills and perceptual motor abilities.

Oxygen therapy during sleep

Many patients on chronic oxygen spend >30% of the night with an Sp,O2 90% while breathing oxygen at the daytime flow rate [17, 18]. If the patient does not have sleep-disordered breathing due to other causes, the administration of oxygen at a flow rate higher than the daytime setting will usually correct nocturnal hypoxaemia.

There is evidence that patients who desaturate only during sleep benefit from nocturnal supplemental oxygen [18]. Increased mortality among patients with nocturnal desaturation and daytime Pa,O2 =8 kPa (60 mmHg) has been demonstrated in retrospective studies [19]. However, well-controlled studies have not shown that use of nocturnal supplemental oxygen alters mortality or clinical course in patients who experience hypoxaemia only during sleep, other than slightly lowering pulmonary artery pressure [20].

Oxygen Therapy During Exercise

Oxygen therapy during exercise decreases dyspnoea and improves exercise tolerance at submaximal workloads [21]. Supplemental oxygen may increase exercise endurance and decrease dyspnoea via a reduction in minute ventilation, dynamic hyperinflation and improved breathing patterns [12, 22].

COPD patients are encouraged to remain active. Many patients with COPD who are hypoxaemic at rest are more hypoxaemic during exertion, while others develop hypoxaemia only during exertion. LTOT is prescribed commonly for both groups, even though studies designed to determine the long-term benefit of oxygen solely for exercise have yet to be conducted.

There is no standardised method for determining exercise oxygen settings. Oxygen settings should be determined while the patient is doing a typical level of exertion, usually walking in a hallway, using the prescribed delivery device (nasal cannula or oxygen conserving device). The goal is to maintain the Sp,O2 >90% when the patient is exercising at a submaximal level equal to or slightly greater than usual exertion in their daily activities.

Appropriate Candidates For Long-Term Oxygen Therapy

Patients whose disease is stable on a full medical regimen, with Pa,O2 7.3kPa (55 mmHg) (corresponding to an Sa,CO2 88%), should receive LTOT.

A patient whose Pa,O2 is 7.3-7.8 kPa (55-59 mmHg) (Sa,CO2 89%) and who exhibits signs of tissue hypoxia, such as pulmonary hypertension, cor pulmonale, erythrocytosis, oedema from right heart failure or impaired mental status, should also receive LTOT.

Desaturation only during exercise or sleep suggests consideration of oxygen therapy specifically under those conditions.

These guidelines are generally accepted and have been adopted by various healthcare systems as reimbursement criteria.

Some gray areas remain, such as patients with adequate Pa,O2 who have severe dyspnoea relieved by low-flow oxygen or patients who are limited in their exertional capacity but improve their exercise performance with supplemental oxygen.

These LTOT indications are summarised in the table 3.

Table 3. – Physiological indications for long-term oxygen therapy (LTOT)

Pa / O2 mmHg Sa / CO2 % LTOT indication Qualifying condition
≤55 ≤88 Absolute None
55-59 89 Relative with qualifier P pulmonale / polycythemia >55% History of oedema
≥60 ≥90 None except with qualifier Exercise desaturation Sleep desaturation not corrected by CPAP Lung disease with severe dyspnoea responding to O2

Pa,O2: arterial oxygen tension; Sa,CO2: arterial oxygen saturation; Right heart failure; CPAP: continuous positive airways pressure; O2: oxygen.

Optimal Medical Regimen

One of the goals of any medical regimen is to optimise V’/Q’ matching as a means of correcting hypoxaemia. This is particularly important during and after an acute exacerbation.

The NOTT [2] incorporated an initial 4-week period in order to optimise medical therapy prior to initiation of oxygen therapy. During this stabilisation phase, nearly 50% of patients who initially qualified for the study according to blood gas criteria alone improved to such an extent on bronchodilators, antimicrobials and corticosteroids (when indicated) that they no longer qualified by blood gas criteria. Consequently, ensuring that the patient is receiving optimal medical management is an important component of the evaluation of need for LTOT. Particular attention should be given to the pharmacological regimen, exacerbation history and the presence of comorbidities that may exacerbate symptoms, e.g. congestive heart failure, sleep-disordered breathing. If the regimen is not optimal but the patient meets criteria, oxygen therapy should be initiated followed by an evaluation in 1-3 months to determine continued need.

One aspect of good medical management is the oxygen therapy itself. Recent reports suggest that oxygen may have a reparative effect by reducing pulmonary artery vasoconstriction, improving V’/Q’ matching and other mechanisms [45]. Accordingly, withdrawing oxygen because of improved Pa,O2 may be detrimental.

Initiating long-term oxygen therapy

When initiating LTOT it is advisable to measure an ABG after breathing room air for 30 min. An ABG is also required to determine the presence of hypercapnia or respiratory acidosis. Pulse oximetry (Sp,O2) is not considered adequate for initiating LTOT. However, oximetry may be used to adjust oxygen flow settings over time.

Administering long-term oxygen therapy

The standard of care for administration of LTOT should be continuous administration (24 h•day-1) with ambulatory capability [2, 46].

Exceptions to continuous administration with ambulatory capability include patients who: 1) are incapable or unwilling to be mobile; 2) require oxygen only during sleep; 3) require oxygen only during exercise; or 4) refuse to use a portable device for ambulation.

Ambulatory oxygen systems

Ambulatory oxygen systems should weigh 10 lb, provide oxygen at 2 L h•min-1 for ≥4-6 h and be packaged so that they can be carried by the patient.

Stationary oxygen

Stationary oxygen may be delivered via a concentrator, compressed gas or liquid. The choice of system will depend upon availability, cost and which portable system is suitable.

Larger portable oxygen systems

Larger portable oxygen systems, such as a steel cylinder on wheels, are suitable for patients who only occasionally go beyond the limits of the stationary delivery system (generally considered to be 50 ft of tubing) [46]. If the patient is not mobile beyond a 150 m radius, an oxygen concentrator is suitable.

Oxygen settings

Oxygen settings should be adjusted for rest, exertion and sleep to meet the individual patient’s needs. See figure 1.

Settings for rest, exertion and sleep

  • Rest : The resting oxygen flow rate can be adjusted, while monitoring oximetry to Sp,O2 ≥90%. ABG should then be used to establish initial Pa,O2 with corroborating oximetry Sp,O2. To insure equilibration, 20-30 min should be allowed after each change in litre flow. For a more accurate reading, the clinician should check the oximeter display for a stable signal and a pulse that is in synchrony with the patient’s heart rate. Fingernail polish, if worn, may need to be removed.
  • Sleep :  The sleep oxygen flow rate can be determined using two strategies: 1) the flow can be increased 1 L•min-1above the daytime resting prescription; or 2) nocturnal polysomnography or nocturnal pulse oximetry can be performed to support a more accurate prescription. If there are signs of cor pulmonale despite adequate daytime oxygenation, the patient should be monitored during sleep to determine the best sleep oxygen setting.
  • Exertion : During exertion, the goal is to maintain Pa,O2 >60 mmHg (8 kPa) or Sa,CO2 >90%. If the patient is using an oxygen-conserving system, titration should be performed while the patient is using that system. This is particularly true during exercise conditions.

Determination of continued need

Standards for continuing oxygen therapy differ depending on whether it is prescribed for the first time during an acute exacerbation or at a time when the patient is relatively stable and receiving optimal therapy [45].

Identification during exacerbation

Some patients with COPD become hypoxaemic during an exacerbation.

Approximately 25-65% of these patients will subsequently heal to the point of not requiring oxygen. When they are clinically stable, this subpopulation should be retested in 30-90 days. If the patient does not meet blood gas criteria at that time, oxygen therapy can be discontinued.

Identification when clinically stable

The majority of patients, who are clinically stable when LTOT is initially prescribed will continue to meet prescribing criteria for LTOT. Some patients, however, experience an improvement in Pa,O2 over time to the point of not meeting the criteria for LTOT. This can occur in stable patients receiving optimal therapy, who have been on oxygen for months or years [45].

When oxygen is withdrawn from these patients, their Pa,O2 begins to decline [45]. The conclusion is that oxygen is reparative (including reversal of hypoxic pulmonary vasoconstriction) and, therefore, should not be discontinued.

Once the need for LTOT has been established in a stable patient on optimal therapy, LTOT is considered to be a lifetime commitment [45].

Physiological indication for long-term oxygen therapy

The physician prescribes oxygen based on physiological findings and clinical judgment. However, LTOT may not be reimbursed unless the patient meets specific physiological criteria. These criteria tend to be similar in most countries. It is therefore essential for the physician to provide appropriate documentation that oxygen is medically necessary and meets the prescribing criteria. The documentation requirements vary widely over the different countries.

Patient Education And Compliance

Patient education and monitoring of compliance are essential to assure the success of LTOT. Many patients harbour fears regarding the therapy. Some patients associate a need for LTOT with profound deterioration rather than prolongation of life and enhancement of quality of life. Some may experience anger or denial and therefore avoid using LTOT. Some patients may avoid using oxygen in public, because they fear reaction of others. Some may regard oxygen as an addictive substance and may therefore avoid its use as much as possible. These and other concerns need to be explored and discussed with the patient and family to provide appropriate rationale and reassurance of the benefits of LTOT.