The MRC study [2] included 87 hypercapnic patients with severe hypoxemia (PaO₂: 40–55 mmHg). The patients received oxygen via concentrator for 15 h/day or no oxygen at all. Portable oxygen cylinders were not provided and those patients who continued smoking were not excluded. The mortality at 3 years was 45.2 % in the oxygen-treated group and 66.7 % in controls. The MRC study showed that LTOT when given for greater than 15 h/day significantly improved survival in patients with hypoxic COPD. The NOTT study [3] enrolled 203 patients with stable hypoxemia (PaO₂ ≦ 55 mmHg or 55 < PaO₂ ≦ 59 mmHg in the presence of cor pulmonale, hematocrit ≧ 55 %, or electrocardiographic evidence of P pulmonale) on two measurements over a 3-week exacerbation-free period. The patients received continuous or nocturnal oxygen and were also followed for a period of 3 years or until death. The continuous oxygen flow rate was sufficient to increase PaO₂ to 60–80 mmHg, with flow rates increased by 1 L/min during sleep and exercise. The mortality rates at 24 months and 3 years were 27 and 32 % for the continuous group and 41 and 50 % for the nocturnal group, respectively. The NOTT study demonstrated a significant survival advantage in the continuous oxygen group, in whom the average oxygen usage was 18 h/day, compared with the nocturnal oxygen only group. Also, the survival advantage was prominent in patients with carbon dioxide retention and also present in patients with relatively poor lung function, low mean nocturnal oxygen saturation, more severe brain dysfunction, and prominent mood disturbances. Continuous O₂ therapy also appeared to benefit patients with low mean pulmonary artery pressure (Ppa) and pulmonary vascular resistance (PVR) and those with relatively well-preserved exercise capacity.

In a similar investigation, the result of a 5-year Japanese international survey from 1986, it has been reported that the patients receiving HOT for 5 years showed a significant improvement in cumulative survival rates compared with a patient not prescribed HOT [5]. Figure 11.2 shows the cumulative survival rates in patients with severe hypoxemic pulmonary emphysema who were prescribed HOT (red circle) or not prescribed HOT due to personal reasons although adapted for HOT (black circle) (PaO₂ ≦ 55 mmHg at rest under breathing room air or 55 < PaO₂ ≦ 60 mmHg at rest under breathing room air associated with pulmonary hypertension or with severe hypoxemia (PaO₂ ≦ 55 mmHg) during exercise or sleep). It was also reported that the cumulative survival rate for female patients was better than that for male patients among the patients receiving continuous oxygen therapy [6] which was the same as in the NOTT study [4], however, according to the report from Sweden, female COPD patients adversely showed a lower survival rate compared with males [7].

Acute beneficial effects of oxygen inhalation on dyspnea, exercise tolerance, and pulmonary hemodynamics are well known in patients with moderate to severe COPD, especially in patients who show desaturation below 88 % during exercise [8]. However, there is little convincing evidence from studies to date that LTOT has significant benefits other than on survival. Indeed, the mechanism for the improvement in survival with oxygen therapy still remains unclear, despite the observation of small improvements in some hemodynamic parameters in the NOTT [4].

Cooper et al. [9] have evaluated the effects of receiving LTOT for at least 15 h/day for 12 years in 72 patients (53 male) with a mean age of 60 years presenting with chronic obstructive airways disease and hypoxic cor pulmonale. The mean PaO₂ and PaCO₂ were 46 mmHg and 52 mmHg, respectively. All patients had a PaO₂ ≦ 60 mmHg and 57 patients had a PaCO₂ ≧ 45 mmHg. Pulmonary hemodynamics were measured in 45 patients before and 12–18 months after the induction of LTOT. There were no significant changes in mean Ppa (from 28.3 ± 10.2 (mean ± SD) to 26.1 ± 11.0 mmHg), cardiac output (from 5.9 ± 1.8 to 6.7 ± 2.8 L/min), and total PVR (from 59.2 ± 25.3 to 51.1 ± 24.7 kPa/L/s); overall 5-year survival was 62 %, but the 10-year survival was only 26 %, indicating an acceleration in death rate at 10 years despite continuing LTOT whereas the cumulative survival rates at 5 years were superior to those in the continuous oxygen group of NOTT (53 % in NOTT). Zielifiski et al. [10] also demonstrated the effects of 6 years of domiciliary oxygen therapy on pulmonary hemodynamics in 95 patients with COPD (72 men, 23 women, mean age: 58 ± 9 years, FEV₁: 0.84 ± 0.31 L, PaO₂:55 ± 6 mmHg, PaCO₂: 48 ± 9 mmHg, Ppa: 28 ± 11 mmHg, PVR: 353 ± 172 dyn × s × cm⁻⁵). After initial assessment, all patients were started on a regimen of LTOT. Pulmonary artery catheterization was repeated every 2 years. The mean Ppa in 39 patients fell from 25 ± 8 to 23 ± 6 mmHg, which was no significant difference, and resulted in a small reduction of pulmonary hypertension after the first 2 years followed by a return to initial values and subsequent stabilization of Ppa over 6 years (in 12 patients who completed 6 years of LTOT, Ppa was 25 ± 7 at entry, and 21 ± 4, 26 ± 7, and 26 ± 6 mmHg at 2, 4, and 6 years, respectively; p < 0.01 for 2 vs. 6 years). The long-term stabilization of pulmonary hypertension occurred despite progression of the airflow limitation and of hypoxemia. These findings suggest that the effect of LTOT on pulmonary hemodynamics is temporary and the disturbances of pulmonary hemodynamics may no longer be important determinants of survival.

Neuropsychological function and quality of life (QOL) were assessed in both continuous and nocturnal oxygen therapy groups at baseline and 6 months in the NOTT. Only 42 % of patients showed improvements of the decreased neuropsychological function at 6 months and there were no differences between the continuous and nocturnal groups [11], but a small improvement of the QOL was found in NOTT. Okubadejo et al. [12] examined the relationships between ADL, quality of life, mood state, and airway obstruction in patients using LTOT and in patients not requiring LTOT. They found no significant difference between groups in health status using the St. George’s Respiratory Questionnaire (SGRQ), and the patients using LTOT were less independent in activities of daily living than those not requiring long-term oxygen therapy. On the other hand, Eaton et al. [13] evaluated the effects of LTOT on QOL as the manner of prospective longitudinal interventional study by comparing between the LTOT group fulfilling criteria and commenced on LTOT and the non-LTOT group not fulfilling criteria and continued on standard care.

Significant improvements in QOL assessed by using the Chronic Respiratory Questionnaire (CRQ), total generic Dartmouth COOP Charts, and anxiety domain of the Hospital Anxiety and Depression scale were noted at 2 and 6 months in the LTOT group. Conversely the non-LTOT group demonstrated a progressive decline in QOL. They concluded that the introduction of LTOT to patients with severe COPD fulfilling standard criteria was associated with early significant improvements in HRQL with sustained or further response at 6 months. It is difficult to determine whether the improvements of QOL were due to more than placebo effect because of the lack of placebo, that is, compressed air instead of oxygen. Also, the control subjects were a non-LTOT group enrolled from the subjects not fulfilling criteria for LTOT in the above two studies. The control subjects should be enrolled from the patients adapted to LTOT at random, and different results might be found. We cannot assert that oxygen improves QOL at present. However, daily life is limited by sustained oxygen treatment for LTOT and patients felt uneasy and the QOL decreased. The effects of LTOT may be dependent on the individual style of daily life. If the portable oxygen devices were developed to be smaller, lighter weight, and easy to carry, the effects of LTOT on ADL and QOL would become clear.

Most international guidelines for the management of oxygen therapy in COPD recommended that LTOT should be considered for patients with stable COPD, who show PaO₂ consistently less than or equal to 55 mmHg at rest when awake and breathing room air and for patients with PaO₂ 56–59 mmHg with polycythemia (hematocrit > 0.55) or clinical, electrocardiographic, or echocardiographic evidence of pulmonary hypertension and/or right heart failure [14, 15, 16]. In Japan, for severe hypoxemia, with PaO₂ are ≦ 55 Torr at rest or 55 < PaO₂ ≦ 60 mmHg with severe desaturation during sleep or exercise under breathing room air, continuing in a stable state for more than one month providing enough medical therapy and pulmonary rehabilitation and a doctor judges the necessity, HOT is prescribed. Also, subjects having pulmonary hypertension are adapted for HOT regardless of cause. All of the guidelines generally recommended that oxygen should be used for as many hours of the day as possible, ideally a minimum of 15 h. Górecka et al. [17] examined the benefit of LTOT on survival in COPD patients with moderate hypoxemia. One hundred and thirty-five patients with COPD who showed their PaO₂ were 56–65 mmHg, borderline severe hypoxemia at rest, and advanced airflow limitation (mean FEV₁: 0.83 L) were randomly allocated to a control and LTOT group. The patients were followed every 3 months for at least 3 years or until death. The cumulative survival rate was 88 % at 1 year, 77 % at 2 years, and 66 % at 3 years. No significant differences were found in survival rates between patients treated with LTOT and controls, nor did longer oxygen use (over 15 h/day) improve survival (Fig. 11.3). They concluded that domiciliary oxygen treatment does not prolong survival in patients with COPD with moderate hypoxemia. It has been suggested that LTOT should be prescribed earlier for patients who complain of severe dyspnea on effort where their PaO₂ are ≧ 60 mmHg at rest. However, it has been demonstrated that there was little improvement of dyspnea, QOL, and neuropsychological dysfunction for the patients who complained of dyspnea on effort and showed mild hypoxemia of PaO₂ 71.4 mmHg as mean values at rest [18]. At the present time, HOT is not approved as the treatment for dyspnea on effort if the patient does not show severe hypoxemia.

Most patients with COPD show more hypoxemia (desaturation) and increased Ppa during exercise, and the supplemental oxygen improves desaturation and increased Ppa [19]. Therefore, the oxygen supply needs to be increased in order to maintain the SpO₂ ≧ 90 % during exercise in patients who were prescribed supplemental oxygen at rest. On the other hand, the use of ambulatory oxygen during exercise or with exertion in patients with mild hypoxemia at rest and unfulfilled criteria for LTOT remains controversial. Meta-analysis about acute efficacy of supplemental oxygen revealed that the supplemental oxygen improved dyspnea and exercise tolerance especially in moderate to severe COPD patients with mild hypoxemia at rest [20]. The acute efficacy is dose-dependent and is suggested to be partly related to a reduction in dynamic hyperinflation and reduced breathing frequency [21]. Also, COPD patients with mild hypoxemia at rest, whose SpO₂ decreased lower than 90 % during the 6-min walking test (6MWT) have been demonstrated to have a poor prognosis [22]. The moderate to severe patients with COPD, whose SpO₂ decreased lower than 90 % within one minute after starting the walk, have a tendency to be prescribed supplemental oxygen at an earlier time [23]. There is no large-scale study about the long-term efficacy of ambulatory supplemental oxygen therapy for patients with COPD who show severe hypoxemia only during exercise.

A 12-week, double-blind, randomized crossover study of O₂ versus cylinder-compressed air in 41 COPD patients with breathlessness and mild hypoxemia at rest but showing exertional desaturation (SpO₂ ≦ 88 %) was examined. Improvements were seen in all domains of the CRQ for cylinder oxygen compared with compressed air [24]. Significant improvements were also noted in anxiety and depression and in certain domains of the SF-36. These benefits cannot be predicted by baseline characteristics or acute response supplemental oxygen. On the other hand, Moore et al. [18] examined a 12-week, parallel, double-blinded, randomized, placebo-controlled trial of cylinder air versus cylinder oxygen trial in 143 patients with COPD, having PaO₂ > 60 mmHg at rest breathing room air and moderate to severe exertional dyspnea including 50 patients with exertional desaturation to ≤88 %. The oxygen or compressed air was provided at 6 L/min intranasally for use during any activity provoking breathlessness. However, no significant differences in any outcome were found between groups receiving air or oxygen (Fig. 11.4).

Statistically significant but clinically small improvements in dyspnea and depression were observed in the whole study group over the 12 weeks of the study and resulted from a substantial placebo effect. They concluded that the domiciliary ambulatory oxygen conferred no more benefits than compressed air in terms of dyspnea, quality of life, or function in breathless patients with COPD who do not have severe resting hypoxemia. It has been demonstrated that the inhalation of either oxygen or compressed air induces a relief of breathlessness through the stimulation of nasal receptors by gas flow, however, the precise mechanism is not known [25]. Indeed, in COPD patients with mild hypoxemia at rest, but showing severe desaturation, effectiveness of long-term domiciliary ambulatory oxygen therapy on QOL with more than placebo effect could not be found. However, even though short-term severe hypoxemia has been demonstrated to aggregate systemic inflammation and increase activity of coagulation, the supplemental oxygen prevents exercise-induced oxidative stress in muscle-wasted patients with COPD [26, 27]. Supplemental oxygen during exercise has been recommended in patients in the United States who show mild hypoxemia at rest, but severe desaturation during exercise [28].

According to a review on the effectiveness of adding oxygen to exercise training in comparison to exercise training without oxygen supplementation in patients with COPD, both hypoxemic and nonhypoxemic subjects can exercise longer and have less shortness of breath when using oxygen during an exercise-training program [29]. However, few data have demonstrated the effects of oxygen on daily shortness of breath, daily activity, and QOL. Nonoyama et al. [30] undertook a series of individual randomized controlled trials (N-of-1 RCTs) to measure the effect of oxygen on QOL in 27 COPD patients with transient exertional hypoxemia. Oxygen significantly increased exercise capacity in a 5-min walk test. Among the whole group, neither the CRQ nor the St. George’s Respiratory Questionnaire showed any statistical or clinical differences between oxygen and placebo. This study does not support the general application of LTOT for patients with COPD who do not meet criteria for LTOT. Furthermore, Spielmanns et al. [31] carried out a 24-week training program with progressively increasing loads and compared the influences of oxygen supplementation by a blinded randomized controlled study. Thirty-six subjects with moderate to severe COPD who were not dependent on LTOT trained under supervision for 24 weeks (3 times/week for 30 min/session) with supplementation of oxygen or compressed air at a flow of 4 L/min. Statistically significant improvements were found in QOL, maximal tolerated load during cycling, peak oxygen uptake, and 6-min walk test after 12 weeks of training, but there were no further benefits of supplemental oxygen.

It has been demonstrated that the addition of noninvasive positive pressure ventilation (NPPV) to an exercise training program in severe COPD (mean PaO₂: 65.4 mmHg at rest under breathing room air) produces greater benefits in exercise tolerance and QOL than after training alone, and suggested that domiciliary NPPV can be used successfully to augment the effects of rehabilitation in severe COPD [32]. Heliox breathing, similar to NPPV, unloads the respiratory muscles and relieves both dyspnea and leg discomfort during exercise [33]. This allows COPD patients to exercise longer prior to exhaustion and enhances the physiological training effect. Despite the current findings, the overall cost of the ventilator setup and the gas mixture makes the use of this therapy cumbersome/impractical and/or too expensive to be incorporated into routine pulmonary rehabilitation programs or training at home.