Article of the Month

Effect of palliative oxygen versus room air in relief of breathlessness in patients with refractory dyspnoea: a double-blind, randomised controlled trial

Abernethy AP, et al.

The Lancet 2010; September 4, vol 376: 784-793

Dyspnoea is a subjective sensation that is frequently described by patients as fatigue upon breathing, air hunger, suffocation, choking or heavy breathing. The prevalence of severe dyspnoea has been reported as 65%, 70% and 90% in terminally ill patients with heart failure, lung cancer, and COPD, respectively.

In the oncological population, dyspnoea can be a direct effect of the cancer, an effect of the therapy, or not related to the cancer or therapy. In addition to cancer, patients may suffer from chronic obstructive pulmonary disease (COPD), congestive heart failure, non-malignant pleural effusion, pneumonitis, air flow obstruction, bronchospasm associated with asthma, and/or anxiety. Moreover, dyspnoea may be a clinical expression of severe anemia, overwhelming cachexia and asthenia causing muscle weakness. Many different causes may co-exist in a patient. A given patient’s intensity of dyspnoea is a multidimensional complex resulting from the interaction of factors that integrate the severity of the above underlying pathophysiology with the perception (i.e. cognition, mood, descending inhibitory pathways, fatigue of the muscles of respiration) and the expression of such things as culture and beliefs. The determination of the relative contribution of each factor to the overall symptom experience can only be made by systematic assessment of all relevant factors. Clinical assessment of dyspnoea should always include, in addition to a thorough physical exam, a complete history of the symptom as to temporal onset (acute or chronic), descriptors, precipitating and relieving events or activities, associated symptoms, response to medication or behavioral changes.

Based on this assessment, symptomatic measures in addition to specific treatments aimed at the underlying cancer and/or other pulmonary and cardiovascular diseases are indicated.

There are only a few RCTs that have been carried out in advanced cancer patients to evaluate the role of opioids (enteral, parenteral, nebulized) and O2 therapy in the management of dyspnoea. With respect to COPD patients, no studies evaluated the role of psychotropic drugs in relieving dyspnoea in cancer patients.

In many patients, a progressive increase of the opioid dose in association with a benzodiazepine (e.g., midazolam) seems to be the only possible solution to reduce dyspnoea even though this may produce mild to severe sedation.

Long-term oxygen therapy is indicated for COPD patients with severe hypoxemia (partial pressure of oxygen in arterial blood (PaO2) ≤ 7.3 KPa at rest).; such treatment improves survival, dyspnoea and functional status.

Oxygen is frequently prescribed with symptomatic palliative intent independently of PaO2 and is considered by some to be a standard of care.

Abernethy, et al. carried out an international, multicentre, double-blind, randomized controlled trial to assess the symptomatic effectiveness of palliative oxygen for patients with life-limiting illness, refractory breathlessness, and PaO2 more than 7.3 kPa compared to room air provided via a modified concentrator. The null hypothesis was that oxygen therapy is not superior to room air in this setting.

The participants were recruited from outpatient pulmonary, palliative care, oncology, and primary care clinics in Australia, USA, and UK. Inclusion criteria were: age > 18; PaO2 more than 7.3 KPa, refractory dyspnoea due to life-limiting illness, maximum treatment for underlying disease, presence of dyspnoea at rest or with minimum exercise, on stable medications for 1 week before participation, expected survival of at least 1 month. Exclusion criteria were eligibility for long-term oxygen therapy, a history of hyperbaric respiratory failure with oxygen, anemia, hypercapnia, cognitive impairment, smoking, a respiratory or cardiac event in the previous 7 days.

Eligible patients with PaO > 7.3 KPa assessed by arterial blood gas analysis, were stratified by baseline PaO2 ≤ 9.3 to ≤ 13.3 KPa, and were randomly assigned in a 1:1 ratio to receive oxygen or room air delivered by a concentrator and nasal cannula at 2 L/min for at least 15 hours per day.

The intervention lasted 7 days with patients, the individuals delivering the interventions, investigators and nurses all blinded to the treatment assignments.

Primary outcome : the symptom of breathlessness right now was assessed twice-a-day by means of a self-reported 0-10 NRS (0= not breathless at all, 10= breathlessness as bad as you can image). A 1-point reduction in the intensity was considered a relevant change for all assessments.

Secondary outcomes : 1. average, worse or relief of dyspnoea in the previous 24 h; 2. functional changes; 3. sleep disturbance; 4. drowsiness; 5. nasal irritation or bleeding; 6. anxiety; 7. QoL assessed every day by the McGill QoL questionnaire; 8. performance status; 9. side effects assesses by a 5-point Likert scale.

All the assessments were reported by the patients in a diary starting 2 days prior to an intervention. At the end of the study the patients reported their overall experience with the intervention.

Results: 239 patients took part in the study with various causes of dyspnoea: 120 patients were assigned to oxygen (63% men) and 119 were assigned to room air (60% men), with age of 73 (11%) and 74 (10%) respectively; and the mean (range) of PaO2 (KPa) was 10.3 (7.5-16.3). 93% of patients on oxygen and 83% on air completed all 7 days of assessments. Oxygen therapy did not provide major benefit over room air when administered at 2 L/min by a nasal cannula to patients with PaO2 more than 7.3 KPa. The frequency of side effects did not differ between groups. Severe drowsiness was reported in 10 % of oxygen group compared with 13% of room air group. Six per cent of patients in room air group and 2% of patients in oxygen group reported severe symptoms of nasal irritation, and 1 patient in this latter group had severe bleeding.

Why I choose this article

Dyspnoea is one of the more devastating symptoms in patients with life-limiting illnesses. In an oncological population dyspnoea is considered to be a prognostic indicator of survival (whether alone or in association with other symptoms and/or performance status) because its frequency and severity increase with progression of the disease.

Oxygen is considered by most physicians and nurses a standard of care in patients with dyspnoea.

The results on the role of oxygen therapy in relieving dyspnoea are still conflicting. Oxygen has been found to improve functional capacity in patients with COPD. At present, the only established indication for supplemental oxygen in dyspnoeic cancer patients is when saturation is less than 90%.

However, when there are doubts about the effectiveness of oxygen for symptom relief in a given patient, particularly when oxygen therapy may be a major problem for a simple and rapid home discharge, an “N of 1” study may be conducted. This is accomplished by performing multiple double-blind crossovers between oxygen and air in a single patient. It is possible to determine with great accuracy, in less than one hour, whether a specific patient does or does not benefit from supplemental oxygen.

Reviewed by Dr. Carla Ripamonti (Italy). Dr. Ripamonti is a member of the IAHPC Board and her bio may be viewed at http://www.hospicecare.com/Bio/c_ripamonti.htm