This article requires a subscription to view the full text. If you have a subscription you may use the login form below to view the article. Access to this article can also be purchased.
The beneficial effects of ambulatory home oxygen have been demonstrated since the 1950s, when Cotes and Gibson gave oxygen to ambulatory COPD patients from small portable high pressures cylinders in the United Kingdom. Over the ensuing 7 decades, oxygen has been prescribed to millions of COPD patients in the home setting. Additionally, it is common clinical practice to prescribe supplemental oxygen when chronic hypoxemic respiratory failure not due to COPD (eg, interstitial lung disease, pulmonary hypertension, kyphoscoliosis, and cystic fibrosis) is present or in patients with hypoxemia at hospital discharge following flares of their underlying chronic respiratory disorder, without any substantial evidence. Despite the importance of long-term oxygen therapy in clinical home management, there are many gaps in our current knowledge regarding its mechanisms of action, indications for prescription, and its effects on important patient outcomes. Research conducted in the 1970s and 1980s still provides the basis for clinical decision making and insurance coverage policies regarding long-term oxygen administration. Remarkably, little current research is being conducted to extend our knowledge regarding the indications, mechanisms, and benefits of long-term oxygen therapy. This review will focus on our current knowledge of the end points for supplemental oxygen at home, such as mortality, effects on functional performance, sensation of dyspnea, cognitive function, and quality of life, and highlight areas where future research is needed.
For the last 70 years, oxygen has been utilized as a therapeutic tool to attenuate the detrimental effects of hypoxemia and reverse the complications of hypoxemia in patients with advanced lung disease. It is estimated that over 800,000 patients annually receive long-term oxygen therapy in the United States, at an annual cost of approximately $2 billion.1 Medicare charges for oxygen were $2.2 billion in 2002, with an annual inflation rate of 12–13% per year.2 Ambulatory oxygen therapy has been provided at home for the purposes of reducing mortality, improving functional status, palliating the devastating consequences of dyspnea, and overall improving quality of life.3–6
Oxygen therapy in stable COPD has been utilized for the purposes of normalizing PaO2, reducing ventilatory drive during exercise, and reducing the sensation of dyspnea. The normalization of PaO2 has been the basis of trials with the goal of prolonging survival in patients with COPD and severe resting hypoxemia. If untreated, hypoxia can progress to tissue hypoxia, which has adverse effects on vital organ function. Hypoxemia can induce vasodilatation of the peripheral vascular beds, thereby causing an increase in heart rate and cardiac output and improved oxygen delivery. High levels of erythropoietin due to persistent hypoxemia induce secondary erythrocytosis in an attempt to increase oxygen capacity, which also can cause increased serum viscosity. All of these mechanisms may have detrimental long-term effects on cardiac and hematologic function and lead to pulmonary hypertension, right ventricular dysfunction, and severe secondary polycythemia, all of which can lead to increased mortality. Increasing FIO2 has been reported to reduce chemoreceptor input and minute ventilation, which may be beneficial in COPD patients by reducing lactate production in exercising muscles, improving oxygen delivery to metabolizing tissues, and overall decreasing the development of dynamic hyperinflation. Oxygen may also help to reduce the sensation of breathlessness by providing a relatively rapid reduction in ventilatory drive, which translates into an improvement in the sensation of dyspnea.1
Supplemental oxygen has not only been shown to affect the control of ventilation, but to regulate pulmonary blood flow and modulate gene expression and cellular activity and interaction throughout the body.7–9 These effects suggest that long-term oxygen benefits may exceed the metabolic effects of improved oxygen delivery and have an impact on enhancing remodeling or repair of the lung. Despite the above beneficial aspects of supplemental oxygen in the ambulatory patient, many of the above benefits are not demonstrated in all COPD patients, especially those with less severe levels of hypoxemia, and patient adherence to supplemental oxygen is variable and at times unpredictable. It is unclear whether the lack of oxygen utilization across all patient groups is related to barriers associated with the use of supplemental oxygen, as opposed to a lack of a broad benefit of supplemental oxygen use in the COPD patient population at large.
Additionally, it is a common clinical practice to prescribe supplemental oxygen without any substantial evidence when chronic hypoxemic respiratory failure not due to COPD (eg, interstitial lung disease, pulmonary hypertension, kyphoscoliosis, cystic fibrosis) mirrors the current criteria used in COPD patients. Additionally, many COPD patients and others, as described above, have oxygen therapy prescribed when they remain hypoxemic at discharge from the hospital following an exacerbation of their underlying chronic disease, despite a lack of data that support the short or long-term benefits of oxygen use in these patient groups.
In this paper we will review the benefits of ambulatory home oxygen on mortality, exercise performance, sensation of dyspnea, functional status, and quality of life.
The Effects of Ambulatory Home Oxygen Therapy on Survival
Chronic hypoxemia that leads to development of cor pulmonale portends a poor prognosis with mortality, ranging from 30–100%. Early, non-controlled studies reported a reduction in mortality in patients with COPD, severe hypoxemia, and cor pulmonale with the use of continuous oxygen therapy for periods ranging from 7 to 41 months.3,5 Two carefully conducted randomized controlled trials demonstrated the beneficial effects of long-term oxygen therapy on survival in patients with COPD and severe resting hypoxemia. These 2 landmarks studies were performed in the late 1970s: the Nocturnal Oxygen Therapy Trial (NOTT) and the British Medical Research Council (MRC) Long-Term Domiciliary Oxygen Therapy Trial examined the effects of long-term oxygen treatment on survival and physiologic function in patients with severe chronic bronchitis and emphysema.
The British MRC trial enrolled 87 patients, all younger than 70 years of age, for chronic bronchitis or emphysema and who had irreversible air-flow obstruction (FEV1 range 0.5–0.75 L), severe hypoxemia (PaO2 range 49.4–51.8 mm Hg), carbon dioxide retention (PaCO2 56−60 mm Hg), and a history of cor pulmonale (mean pulmonary artery pressure range 32.3–35.0 mm Hg).10 Those randomized to oxygen were given 2 L/min of oxygen via nasal prongs for at least 15 hours daily. At 5 years of follow-up, 19 of 42 oxygen treated patients died, compared with 30 of 45 control patients who did not receive supplemental oxygen. Mortality was highest in the subgroup of patients that had the highest elevations in baseline PaCO2 and red cell mass. Although mortality was beneficially affected by long-term oxygen use, there were no statistically significant differences in the rate of decrease in FEV1 or PaO2 or increases in PaCO2, red cell mass, or pulmonary artery pressures. However, there was a trend that long-term oxygen therapy prevented a progressive decrease in PaO2 and the rate of increase in pulmonary vascular resistance without provoking further increases in PaCO2.
The NOTT evaluated 203 hypoxemic COPD patients who were randomized to receive either continuous oxygen therapy or 12 hours of nocturnal oxygen therapy.11 All subjects were followed for at least 12 months to determine the effects of oxygen on survival, pulmonary vascular pressures, quality of life, and neuropsychological function. Mortality in the 203 patients was followed for an average of 19.3 months. Twelve month mortality in the nocturnal oxygen group was 20.6%, versus 11.9% in the continuous oxygen therapy group. In the nocturnal oxygen group, 24 month mortality was 40.8%, versus 22.4% in the continuous oxygen treatment group. Hematocrit values were more reduced in patients on continuous oxygen therapy, versus those randomized to nocturnal treatment, as was pulmonary vascular resistance. However, no effect of continuous versus nocturnal oxygen therapy was found on other physiologic parameters, such as FEV1, mean pulmonary artery pressures, lung volumes, or arterial blood gases.
It should be noted, however, that, although both of the above studies showed improved survival, there were some important differences between the 2 studies. MRC patients tended to be more ill and had more evidence suggesting hypercapnia or cor pulmonale. A substantial number of subjects in the MRC study continued to smoke even after enrollment into the trial (27% in the control group and 44% in the placebo group). The MRC study found no effect of nocturnal oxygen therapy versus no oxygen therapy on physiological variables, while NOTT found greater decreases in hematocrit and pulmonary vascular resistance in those who used continuous oxygen therapy versus nocturnal use only.
Despite these differences between the 2 trials, however, the results of these 2 trials indicate that the duration of use of supplemental oxygen is an important factor in improving survival in COPD patients with severe resting hypoxemia (Fig. 1). Median survival in those using oxygen 18 hours per day were approximately 2-fold longer than in those receiving no oxygen therapy. Subsequent uncontrolled studies examining patient populations similar to the NOTT and MRC studies that were treated with oxygen have produced similar results.
In contrast to the NOTT and MRC studies, 2 studies have reported that supplemental oxygen use in COPD out-patients with mild to moderate resting hypoxemia has no effect on survival.12,13 Both of these studies did not use other indices of disease severity, and enrolled only 81 subjects who were followed for a maximum of 85 or 36 months. The average daily use of oxygen in the 2 studies was approximately 13.5 hours in those randomized to receive oxygen. This duration may be inadequate because COPD patients receiving long-term oxygen therapy have been reported to have an increase in pulmonary vascular resistance when oxygen was stopped for as little as 3 hours a day.14 Although neither of the studies was powered to examine the effect of oxygen in mild to moderate resting hypoxemia on survival, the similarities of outcomes between the treated and control groups suggest that patients with less severe COPD fail to derive any survival benefit from long-term oxygen therapy.12,13
Despite the above data, clinicians continue to prescribe oxygen therapy to patients with severe COPD, even when severe resting hypoxemia is absent. In an analysis of patients enrolled in the National Emphysema Treatment Trial, 33.8% of the patients with severe emphysema and FEV1 < 45% of predicted without severe resting hypoxemia received continuous oxygen therapy.15 The National Emphysema Treatment Trial patients receiving continuous oxygen had worse dyspnea and poorer quality of life and more frequent oxygen saturation with exercise, compared to a comparable National Emphysema Treatment Trial subgroup not receiving oxygen. In those patients who utilized oxygen therapy continuously, a higher mortality was observed, in contrast to a similar comparison group that did not use oxygen (Fig. 2). These data illustrate a divergence between evidence-based studies reporting those who are most likely to benefit from continuous oxygen, compared to the characteristics of patients who are prescribed oxygen in current clinical practice.
Effect of Long-Term Oxygen on Physiological and Functional Outcomes
Pulmonary hypertension is a common complication of severe COPD and chronic hypoxemia, and is associated with increased mortality, risk for hospitalization, and duration of hospitalization that is independent of the degree of air-flow obstruction. Although it has been suspected that the increased life expectancy with supplemental oxygen in COPD patients with severe hypoxemia is attributable to an improvement in pulmonary hemodynamics, data supporting that notion are weak and contradictory. The NOTT reported that oxygen therapy in hypoxemic COPD patients was associated with favorable but minimal improvements in the magnitude of pulmonary hypertension at rest and during exercise.16 Additionally, NOTT data suggested that improvements in pulmonary hemodynamics secondary to oxygen supplementation were enhanced in the group of patients who used oxygen continuously, compared to those who used it on an intermittent daily basis.16 However, improvements in survival in NOTT patients receiving continuous oxygen, compared to those who received only nocturnal oxygen, were not attributable to changes in pulmonary hemodynamics, and the acute response to oxygen administration on pulmonary hemodynamics did not predict the long-term effects of oxygen on survival.
Three subsequent studies have examined the effect of oxygen on the pulmonary circulation and have attempted to determine: whether the effect of long-term oxygen therapy and pulmonary hypertension is maintained longer than 6 months in COPD; whether the acute effects of oxygen administration on pulmonary artery pressures predicts survival in hypoxemic COPD patients treated with long-term oxygen; and the effect of acute withdrawal of supplemental oxygen on pulmonary artery pressures in COPD.
Weitzenblum et al studied 16 patients with severe COPD (FEV1 0.89 ± 0.28 L) who were hypoxemic at rest (PaO2 50.2 ± 6.6 mm Hg) and moderately hypercapnic (PaCO2 51 ± 6.4 mm Hg) by conducting 3 consecutive right heart catheterizations before and after oxygen therapy.17 Oxygen therapy was provided 15–18 h/d and prescribed at flow rates based on usual criteria. After long-term oxygen administration was initiated, with subsequent follow-up at an average of 1/2–2 years, pulmonary artery mean pressure decreased from 28.0 ± 7.4 mm Hg to 23.9 ± 6.6 mm Hg (P < .005). Following the administration of supplemental oxygen, a yearly decrease in mean pulmonary artery pressure of 2.15 ± 4.4 mm Hg was observed. These data were interpreted as demonstrating that long-term oxygen therapy has a beneficial effect on reducing pulmonary artery pressures for a period greater than 6 months.
Sliwinski and colleagues studied the acute effects of supplemental oxygen on pulmonary hemodynamics as a predictor of long-term outcome in 46 severe COPD patients who otherwise qualified for long-term oxygen administration.18 Thirty-nine of the subjects had a decrease in mean pulmonary artery pressure ≤ 5 mm Hg in response to an acute administration of oxygen, and were called oxygen non-responders. In 7 patients, mean pulmonary artery pressure decreased > 5 mm Hg, and they were labeled oxygen responders. Patients were then prescribed long-term oxygen therapy by conventional criteria for 2 years or until death. Two-year survival rate was 69% in the non-responders and 50% in the responders. From these data it appears that the long-term use of oxygen may significantly improve survival, despite the lack of an acute oxygen effect on reducing pulmonary artery pressures.
Selinger et al performed repeated right heart catheterizations in 20 stable hypoxemic COPD patients while receiving supplemental oxygen, and then after interrupting the use of oxygen to determine whether reductions in pulmonary artery pressures were related to the daily duration of oxygen use.14 When oxygen therapy was withdrawn, the mean PaO2 decreased from 74 ± 4 mm Hg to 55 ± 2 mm Hg. Pulmonary artery pressures significantly increased at rest and nearly doubled with exercise. Discontinuation of oxygen reduced the stroke volume index and increased the pulmonary vascular index by approximately 30%. These alterations occurred gradually and reached steady state levels 2–3 hours following oxygen discontinuation. These data suggest that an interruption in oxygen supplementation for several hours a day in chronically hypoxemic COPD patients can result in an elevation of pulmonary artery pressures at rest and during exertion.
A recent controlled trial has reported that a 3-month period of nitric oxide and oxygen inhalation leads to a sustained improvement in resting pulmonary hemodynamics in severe COPD patients with secondary pulmonary hypertension, compared to oxygen treatment alone.19 Whether a combination of nitric oxide and oxygen treatment leads to an improvement in survival in severe COPD patients with secondary pulmonary hypertension is unknown at present.
In summary, the above data show that inhalation of supplemental oxygen with hypoxemic COPD produces favorable but minimal changes in the magnitude of pulmonary hypertension at rest and during exercise. Furthermore, it also suggests that favorable changes in pulmonary hemodynamics secondary to oxygen supplementation are enhanced in patients that receive oxygen continuously, compared to those who receive it on an intermittent basis. Improvements in survival in COPD patients receiving continuous versus nocturnal oxygen therapy is difficult to attribute solely to changes in pulmonary hemodynamics, based on the available data, and the acute response to administration of oxygen on pulmonary hemodynamics does not appear to predict the long-term effects of oxygen on survival. Further studies are required to determine the causal relationship of oxygen supplementation and pulmonary hemodynamics on survival and which subgroup of patients with pulmonary hypertension and hypoxemia in COPD are most likely to benefit from long-term oxygen administration.
Hematocrit values in COPD reflect the balance between chronic hypoxemia stimulating erythropoiesis, and comorbidities and chronic inflammation suppressing erythropoiesis. In NOTT, patients with a high pulmonary vascular resistance and hematocrit showed the highest mortality.11 After 6 months of supplemental oxygen, a significant reduction in pulmonary vascular resistance and hematocrit was observed, but the long-term effects of this response on mortality in NOTT were not demonstrated. A recent retrospective, observational study evaluated the prognostic value of hematocrit in 2,524 severe COPD patients who received long-term oxygen therapy.20 Lower hematocrit values at baseline before implementation of long-term oxygen therapy was associated with worse survival. Celli et al similarly reported that lower hematocrit values in COPD patients were associated with increased mortality.21 At present, the prognostic value of low hematocrit values in severe COPD and the effects of supplemental oxygen on affecting the hematocrit or erythropoietin production are unclear and require further study.
Several features of COPD may contribute to cognitive dysfunction, including hypoxemia, cerebrovascular and cardiovascular disease, and limited physical activity that leads to an increase in social isolation. In NOTT, patients who received 6 months of long-term oxygen therapy were found to have improved alertness, motor speed, and hand grip but not improved emotional status or quality of life.22 An additional study in COPD patients after 3 months of long-term oxygen therapy demonstrated a slight positive effect of oxygen use on cerebral blood flow, autonomic function, and neuropsychologic status.23 A recent study of cognitive function in 1,202 COPD patients examined potential risk factors for cognitive impairment in comparison to age-matched normal controls.24 COPD was associated with an increased risk of cognitive impairment (odds ratio 2.42, 95% CI 1.043–6.64). In the COPD cohort, a low baseline oxygen saturation was related to an increased risk of cognitive impairment (odds ratio for oxygen saturation ≤ 88% 5.45, 95% CI 1.014–29.2, P = .048). Regular use of supplemental oxygen decreased the risk for cognitive impairment (odds ratio 0.14, 95% CI 0.07–0.27, P < .001). Despite the important implications of cognitive dysfunction in COPD patients, the effects of supplemental oxygen on improving cognitive dysfunction is understudied at the present time and merits further investigation.
Dyspnea is a common and distressing symptom found in patients with COPD, especially in the advanced stages of the disease. Studies that have examined the use of oxygen therapy for treating breathlessness in patients with COPD at rest are limited in number and include small numbers of patients, with contradictory outcomes. More studies are found examining the short-term use of supplemental oxygen to decrease the sensation of breathlessness during exercise, and most studies show a reduction in dyspnea and an improvement in exercise endurance with the use of supplemental oxygen. Trials of long-term oxygen that measure breathlessness are few in number and report contradictory effects of long-term supplemental oxygen on reducing breathlessness.
A recent report of an expert working group of the Association of Palliative Care provides the following summary of evidence on the use of oxygen on decreasing breathlessness in COPD patients25:
There is evidence for and against using oxygen to palliate breathlessness at rest.
The majority of patients using enriched oxygen during exercise experienced less breathlessness at equivalent levels of exercise, when compared to just inspiring room air oxygen.
There is no evidence that pre-oxygenation reduces breathlessness during exercise.
There is recent evidence that using oxygen may speed recovery from breathlessness, when given before or after exercise.
Recent data suggest that the effects of ambulatory oxygen on quality of life over a longer term cannot be predicted from the patient's baseline characteristics or their acute/short-term response to oxygen therapy. Even when patients experience an acute short-term favorable effect with the use of oxygen therapy, a substantial proportion do not continue to use ambulatory oxygen because of poor tolerability.
Quality of Life
Improving quality of life is an important clinical outcome for patients with severe COPD; whether long-term oxygen administration improves quality of life in hypoxemic COPD patients is unclear at present. Quality of life was not addressed in the MRC trial, whereas in the NOTT a generalized measure, the Sickness Impact Profile was measured and demonstrated similar outcomes after 6 months of continuous or nocturnal oxygen treatment.11 Only 2 longitudinal studies have reported the effects of long-term oxygen on quality of life, perhaps because of ethical reasons or the difficulty of incorporating a placebo-controlled arm into the study design.5 Furthermore, it has been reported that the potential restriction on mobility imposed by using the oxygen device may be an important factor that counterbalances any improvement by using supplemental oxygen on quality of life.26,27 Quality of life has been reported not to improve when patients received oxygen therapy for 6 months while using an oxygen concentrator, whereas other investigators have shown improved quality of life in patients who received liquid oxygen treatment, versus those who received oxygen via concentrators or small oxygen portable cylinders.27 In contrast, however, in a 1-year, randomized crossover trial, 24 COPD patients randomized to home oxygen therapy with an oxygen concentrator only, home oxygen concentrator plus as needed ambulatory oxygen, and home oxygen concentrator plus ambulatory compressed air treatment regimens showed no effect of additional ambulatory oxygen on quality of life, 6-min walk tests performance, or the daily duration of exposure to oxygen therapy.28
An important question is whether improvement of quality of life relates to either the severity or chronicity of hypoxemia and the pattern of supplemental oxygen use in COPD patients. Eaton et al reported that short-term use of ambulatory oxygen produced significant improvements in quality of life in COPD patients who demonstrated substantial exertional desaturation.29 However, the correction of oxygen desaturation was not predictive of the acute or longer term improvement in quality of life. Additionally, the same authors found that long-term oxygen administration in severe COPD patients was found to produce improvements in quality of life that were sustained/or further improved at 6 months.30 However, Nonoyama and colleagues reported no effect of ambulatory oxygen on quality of life measured by the Chronic Respiratory Questionnaire or St George's Respiratory Questionnaire, compared to standard therapy.31
It is important to stress that in a disease like COPD, which has high morbidity and impaired quality of life, with very few viable limited therapeutic options, the exploration of any therapy that may palliate the disease, regardless of a mortality effect, is beneficial. The contradictory nature and sparseness of the above data illustrate the need for further research in this important area.
There are no current data to support the notion that long-term oxygen therapy may attenuate the decline in FEV1 or that FEV1 may be an important selection criterion to characterize the most appropriate candidates for long-term oxygen therapy. This is not surprising, since many of the comorbidities associated with COPD, such as pulmonary hypertension and cor pulmonale, polycythemia, congestive heart failure (CHF), stroke, and cerebrovascular disease, may have their negative consequences reversed or attenuated by the use of supplemental oxygen. The likelihood that the severity of air-flow obstruction may independently predict differential response to long-term oxygen therapy does not appear to be a plausible hypothesis.
Exercise-Induced Oxygen Desaturation.
Another group of COPD patients who may benefit from ambulatory oxygen therapy are those who are normoxemic or only mildly hypoxemic at rest but exhibit oxygen desaturation during physical activity. This patient group undoubtedly highlights the most controversial application of chronic oxygen therapy in COPD. Continuous oxygen therapy in this patient population has been shown to improve outcomes related to exercise performance and endurance capacity, but has not been shown to improve survival. Exertional desaturation is fairly common in advanced COPD, and Medicare and other healthcare providers pay for oxygen use in patients who exhibit only exertional desaturation.32 In both normal individuals and patients with COPD, supplemental oxygen improves exercise performance. This improvement has been thought to result from increasing oxygen transport, enabling greater utilization of oxygen by exercising muscles, avoiding or delaying the onset of inspiratory muscle fatigue, an improvement in right ventricular function, or a decrease in ventilatory work load because of decreased airways resistance.
The prevalence of exertional desaturation in COPD is unclear and depends on the characteristics of the population studied as well as the exercise protocol used to elicit oxygen desaturation. Soguel Schenkel et al reported in a group of 30 moderate to severe COPD patients (meaning FEV1 37% of predicted, resting PaO2 88 mm Hg) a greater number of desaturations during walking (13.1 desaturations/h), washing (12.6 desaturations/h), and eating (9.2 desaturations/h) than at rest (5.3 desaturations/h).33 More desaturations occurred during the day, compared to the night (8.6 vs 6.8 desaturations/h, respectively). Performance during a 6-min walk test may predict the presence of important desaturations while performing the activities of daily living. Garcia-Talavera et al showed that COPD patients with moderate hypoxemia at rest did not have important desaturations during daytime activities if they did not desaturate during the first 3.5 min of a 6-min walk test.34
Predicting Which Patients May Have Exercise Desaturation.
Patients with predominantly an emphysematous phenotype and a lower diffusion capacity are more likely to desaturate during exertion, compared to patients with a similar degree of air-flow obstruction but an airway predominant rather than emphysema predominant phenotype. In 48 patients with COPD and resting PaO2 > 55 mm Hg at rest, 40% desaturated doing incremental lower extremity cycle ergometry testing. FEV1 and diffusion capacity were lower in those who desaturated, compared with individuals who did not desaturate (0.89 L and 7.1 mL/min/mm Hg vs 1.44 L and 15.3 mL/min/mm Hg, respectively).35 No patient with a diffusion capacity of > 55% of predicted exhibited desaturation. Approximately 75% of patients with a diffusion capacity or FEV1 < 35% of predicted desaturated during lower extremity cycling.
Resting oxygen saturation appears to be a useful tool to predict which patients may desaturate during exertion. Knower et al demonstrated in 81 COPD patients that approximately 50% of patients with a resting arterial oxygen saturation (SaO2) ≤ 95% experience desaturation during the 6-min walk test, in contrast to only 16% with a resting SaO2 > 95% who desaturate with exertion.36 In a subgroup of 70 patients with diffusion capacity measurements, no patient with a diffusion capacity > 36% and resting SaO2 > 95% desaturated during exercise. Moreover, Hadeli et al found the sensitivity and specificity for predicting exercise-induced desaturation at approximately 75% when the threshold for diffusion capacity was kept at 59.7% of predicted.37 Patients who perform an incremental shuttle walk test are more likely to desaturate, in comparison to those performing a 6-min walk test.38 Lewko et al reported that the mean drop in oxygen saturation was −4.6% in 50 stable COPD patients performing a shuttle walk test, compared to −2.8% in the same individuals performing a 6-min walk test.38
Exercise Desaturation and Mortality.
A limited number of studies, comprised of small patient numbers, suggest that transient exertional desaturation may be associated with increased mortality.39–42 This has some biologic plausibility, since exercise-induced desaturation has been previously shown to correlate with the severity of pulmonary vascular disease in patients with only mild-moderate resting hypoxemia.43 However, limited information exists as to whether supplemental oxygen alters the course of the disease or prolongs survival. Overall, resting oxygen levels appear to have greater predictive power than exercise-induced desaturation in predicting the course of the disease or response to long-term oxygen therapy.
Effects of Oxygen Therapy on Exercise Tolerance.
The most troubling symptoms that afflict COPD patient are dyspnea and impaired exercise tolerance. Several studies have shown the beneficial effects of supplemental oxygen on improving exercise performance in those who are hypoxemic at rest. Cotes and Gibson were among the first to assess the effects of supplemental oxygen on improving exercise performance in COPD patients and chronic respiratory failure.44 They found that breathing 30% oxygen significantly increased the time that subjects could walk on a treadmill, compared with breathing room air. Several other investigators confirmed their findings that COPD patients who were hypoxemic at rest had significant improvement in exercise performance when provided supplemental oxygen. Davidson et al showed a dose-dependent effect of supplemental oxygen on improving exercise performance in 17 patients; supplemental oxygen at a flow of 2 L/min increased constant lower extremity cycling work load duration by 51%, compared with room air, and by 80% with 4 L of oxygen per minute, and by 80% with 6 L of oxygen per minute.45 Legget and Flenley found that patients suffering from resting hypoxemia could increase exercise tolerance assessed by distance walking 12 min while breathing oxygen, but that this improvement in exercise tolerance was negated when subjects were forced to carry their portable oxygen supply.46 The improvement in walking distance was restored when supplemental oxygen use during the 12 min walk test was provided by wheeling the oxygen source on a light-weight trolley.
Determining which patients suffering from COPD may show improved exercise with supplemental oxygen has been somewhat problematic. Cotes and Gibson suggested that patients who benefited the most from supplemental oxygen improving exercise capacity had the largest venous admixture at rest, the greatest decrease in PaO2 during exertion, and the greatest increase in PaCO2 during exercise.44 Bradley and colleagues studied 26 severe COPD patients exercising on a treadmill while breathing supplemental compressed room air or oxygen.47 Although they found that treadmill exercise endurance time was significantly increased while inspiring supplemental oxygen, no relationship was found between increased exercise endurance and the degree of hypoxia, hypercarbia, or acidosis during exercise, or the changes in these parameters while breathing enriched oxygen, compared to breathing air.
Several small studies have suggested that supplemental oxygen enhances exercise performance in patients who are normoxemic at rest but desaturate with exertion. Voduc et al compared the effects of supplemental oxygen on exercise capacity in 16 hypoxemic COPD patients before and after participation in a pulmonary rehabilitation program.48 Supplemental oxygen therapy improved exercise duration by 75 seconds before participation in the pulmonary rehabilitation program, and by 153 seconds following pulmonary rehabilitation. A double-blind randomized placebo controlled trial was conducted in 20 stable COPD patients to assess oxygen desaturation during activities and to assess the short-term effects of supplemental oxygen.49 Patients underwent three 6-min walk tests in random sequence: a baseline 6-min walk, and 6-min walk tests while breathing compressed air or oxygen. In patients who exhibited oxygen desaturation, supplemental oxygen, compared to compressed air inhalation, resulted in a 22% increase in the distance walked and a 36% reduction in the magnitude of dyspnea. In patients who did not desaturate, oxygen administration reduced the magnitude of dyspnea by 40%, but the distance walked remained unchanged. Emtner et al studied whether non-hypoxemic COPD patients undergoing exercise training while breathing supplemental oxygen could achieve higher intensity and therefore improved exercise capacity, compared to patients breathing just room air.50 In a double-blind trial conducted in 29 non-hypoxemic COPD patients, the oxygen training group increased training work rate more rapidly than the air training group. The mean ± SD work rate during the last week was 62 ± 19 W in the oxygen training group vs 52 ± 22 W in the air training group (P < .01). After training, endurance and the constant work rate increased more in the oxygen training group (14.5 min) than the air training group (10.5 min, P < .05). At exercise isotime, breathing rate decreased 4 breaths/min in the oxygen training group and 1 breath/min in the air training group (P = .001).
Other investigators have found limited or no benefit on exercise performance while using supplemental oxygen in COPD patients who are non-hypoxemic at rest. Eaton et al conducted a 12 week double-blind randomized crossover trial in 50 severe COPD patients (mean FEV1 25.9 ± 8.0% of predicted, resting PaO2 9.2 ± 1.0 kPa, SaO2 after 6-min walk 82 ± 5.4%), and 68% of the patients increased their 6-min walk distance ≥ 54 m or had a decreased Borg dyspnea score of ≥ 1 unit.29 All measures on the Chronic Respiratory Questionnaire and levels of anxiety and depression determined by the Hospital Anxiety Depression Scale and general health, physical, and emotional roles on the Short Form-36 improved with supplemental oxygen. However, the improvements in patient-reported quality of life outcomes did not correlate with oxygen's effects on the 6-min walk distance or the Borg dyspnea score. Despite measurable beneficial effects, nearly half of the responders chose not to use continuous supplemental oxygen on the long term.
In another crossover study in 36 COPD patients (FEV1 0.9 L, resting PaO2 69 mm Hg) there was no beneficial effect of supplemental oxygen on health status, exercise tolerance, or dyspnea, when receiving either 6 weeks of portable oxygen or air.51 In an additional crossover randomized trial comparing portable oxygen versus air in 9 patients with a resting PaO2 < 60 mm Hg, no differences in 6-min walk distances were found.52 In a single blind trial in 20 patients, no significant differences in 6-min walk distance were found after 8 weeks of treadmill exercise training using either supplemental oxygen or compressed air.53 Rooyackers et al examined the effect of inhaled oxygen or air on exercise performance and quality of life during 10 weeks of in-patient pulmonary rehabilitation in 24 COPD patients with exertional desaturation to < 90%.54 They reported that oxygen had no effect on quality of life or exercise performance. In a prospective randomized placebo controlled trial of air versus cylinder oxygen provided at 6 L/min intranasally over 12 weeks in 143 moderate to severe COPD patients, no significant differences in any outcome were found between the groups receiving air or oxygen therapy.55 Finally a meta-analysis of 3 small studies with the purpose to determine the effectiveness of adding oxygen for exercise training, in comparison to exercise training without oxygen supplementation, in COPD reported that patients may exercise longer and have less shortness of breath when using oxygen during an exercise training program; however, it was not possible to determine which individuals should use oxygen during exercise training because of the small number of subjects involved and the lack of characterization of those most likely to benefit.56 Additionally, it was impossible to tell whether oxygen inhalation had any significant effect on shortness of breath while subjects performed the activities of daily living.
Mechanisms by Which Oxygen Therapy May Improve Exercise Tolerance.
Identifying the mechanisms responsible for producing the improvements in exercise endurance with inspired oxygen has also proven to be problematic (Table 1). Although several investigators have found that inspiring enriched oxygen during exercise reduces the perception of breathlessness, the mechanisms for this effect have remained a matter of debate. Some investigators have attributed the reduction in perceived breathlessness to the effect of cold gas stimulating the nasal mucosal receptors rather than to an improvement in oxygenation levels. Swinburn and co-workers demonstrated in 12 patients with severe COPD and resting hypoxemia that inspiration of 28% oxygen, in contrast to compressed air, was associated with a significant improvement in SaO2 (85.1+ 2.3% vs 93.1 ± 1.4%) and a reduction in breathlessness.57 Additionally, when the patients breathed oxygen, minute ventilation was significantly reduced, when compared to when they breathed air. These investigators interpreted their data as showing that the increase in arterial oxygen associated with breathing enriched oxygen resulted in a reduction in minute ventilation, primarily due to a reduction in tidal volume.
Besides a reduction in minute ventilation while inspiring oxygen, other investigators have attributed oxygen-related exercise improvements to increased peripheral oxygen delivery, an improvement in pulmonary hemodynamics and right ventricular function, or to improvements in respiratory and non-respiratory skeletal muscle function. Morrison and Stovall found that hypoxemic COPD patients who had improved exercise capacity with supplemental oxygen had significant increases in oxygen delivery during exercise, which was associated with increases in both cardiac output and oxygen content, in contrast to those who had no improvement in exercise endurance.58 Similar data suggesting a beneficial effect of supplemental oxygen on improving exercise capacity due to an improvement in central pulmonary hemodynamics and right ventricular function were demonstrated in the NOTT.16 In the NOTT, breathing oxygen was associated with a reduction in cardiac index, right ventricular stroke work index, and mean pulmonary artery pressures at rest. During cycling exercise while breathing oxygen, in contrast to breathing room air, reductions in right atrial pressure, mean pulmonary artery wedge pressure, pulmonary vascular resistance, and stroke work index were also reported. Although the above studies demonstrate that oxygen may beneficially impact pulmonary hemodynamics both at rest and with exercise, the correlation of improvements in central pulmonary hemodynamics and cardiac performance with increased exercise endurance while breathing supplemental oxygen is unclear at the present time.
Improvements in ventilatory muscle function have also been proposed as an additional mechanism of improved exercise endurance while inspiring supplemental oxygen. Bye et al reported, in a group of severely obstructed patients, that supplemental oxygen increased exercise performance, which was accompanied by reduction in minute ventilation and breathing frequency, in contrast to breathing air.59 In some of these individuals there was a delay in the appearance of the electromyographic pattern of diaphragm fatigue. Criner and Celli reported, in a crossover prospective randomized trial, that inspiring supplemental oxygen during exercise altered ventilatory muscle recruitment.60 In 6 subjects with severe COPD and mild hypoxemia at rest (mean PaO2 66 ± 6.2 mm Hg), measurement of transdiaphragmatic pressures during exercise suggested a redistribution of ventilatory muscle recruitment that enabled more dynamic work to be done by diaphragm and less by the accessory respiratory and abdominal expiratory muscles while breathing enriched oxygen, compared to air. Subsequent work by others suggest that exercising with oxygen in COPD results in a reduction in minute ventilation and an increase in expiratory time that results in less dynamic hyperinflation and thereby a much more favorable environment that enhances diaphragm mechanical action and reduces accessory respiratory muscle work.1,61
Supplemental oxygen may also have important cellular effects on skeletal muscle performance. Acute correction of hypoxemia with supplemental oxygen in COPD patients has been reported to decrease the ratio of organic phosphate to phosphocreatine and increase the resynthesis rate of creatine phosphate.62 Additional investigators have shown that an improvement in skeletal muscle energy metabolism may occur during long-term oxygen treatment in patients with severe COPD and chronic hypoxemia.63
The inconsistent results reported by studies examining the effects of supplemental oxygen on exercise endurance can in large part be explained by limitations in quality of the research that has examined this problem to date. A recent systematic review of randomized controlled trials that examined the short-term benefits of ambulatory oxygen in COPD suggests that ambulatory oxygen improves exercise performance in COPD; however, the clinical importance of the magnitude of the treatment effect is unclear at the present time.64 Most of the reported data come from studies that have small patient numbers and are conducted in variable COPD patient populations with a variety of different exercise outcomes being used to assess the effects of supplemental oxygen that is provided at different doses and for variable durations. Future quality research that takes these design issues into consideration are needed to determine the place of supplemental oxygen in determining the efficacy and mechanisms of this potentially important therapy.
Short Burst Oxygen Therapy.
Short burst oxygen therapy describes brief intermittent uses of oxygen before or after exertion.65 It has been used in several ways, such as pre-oxygenation before exercise, for the treatment of breathlessness during recovery from exercise, or for the control of breathlessness at rest. Short burst oxygen therapy is predominantly utilized by COPD patients who are not hypoxemic at rest or with exercise, and is usually provided by oxygen cylinders rather than oxygen concentrators. Although early studies suggested that short burst oxygen reduced dyspnea and increased 6-min walk distance, subsequent studies have not demonstrated a reduction in breathlessness or improved walk test performance.61,66–71 A randomized single-blind crossover study showed that breathing oxygen at 40% concentration, but not air, for 15 min after exercise in 18 moderately severe COPD patients decreased the time for recovery of dynamic hyperinflation by an average of 6.61 min.61 Neither the maximal level of dyspnea nor duration of breathlessness was improved by short burst oxygen therapy. In a recent prospective randomized trial, 30 moderate to severe COPD patients with FEV1 < 40% of predicted and resting oxygen saturation ≥ 93% were given 4 L of oxygen via a simple face mask, 4 L of air from a simple face mask, air from a fan, or no intervention.72 Oxygen therapy had no discernible effect on Borg dyspnea scores during recovery from exercise or shortening the recovery period of dyspnea. The benefits of short burst oxygen therapy remain to be determined.
The comorbidities associated with COPD are being increasingly recognized as important contributors to the morbidity and mortality of the disease. Clinical practice guidelines do not provide any guidance regarding the use of long-term oxygen therapy in patients with COPD and important comorbidities. In fact, the body of research to date has extensively excluded patients with important comorbidities from participating in studies that have examined the use of long-term oxygen therapy in COPD. A recent retrospective study demonstrated that body mass index < 25 kg/m2 and the presence of comorbidities are predictors of all-cause respiratory mortality in COPD patients treated with long-term oxygen therapy.73 It will be important in the future to study the true impact of long-term oxygen therapy on all-cause mortality in complex COPD patients. Although it is well known that cardiovascular and cerebrovascular comorbidities increase mortality risk in COPD, these comorbidities were never taken into account when examining the outcome of long-term oxygen therapy in prior studies of COPD patients. At the current time there are insufficient data to determine whether long-term oxygen therapy reduces cardiovascular/cerebrovascular and/or metabolic morbidity and mortality in hypoxemic COPD patients.
Long-Term Oxygen in Other Lung Diseases
Interstitial Lung Diseases
Hypoxemia is common in patients with interstitial lung diseases, especially those with idiopathic pulmonary fibrosis. Oxygen therapy may have the potential to either prolong or shorten survival in patients who suffer from idiopathic pulmonary fibrosis or usual interstitial pneumonitis. Hypoxemia at rest and desaturation with exertion in patients with idiopathic pulmonary fibrosis have been reported to be variables that predict worsened prognosis.74,75 With progression of the disease, hypoxemia may contribute to the development or worsening of pulmonary hypertension, which can lead to right heart dysfunction or cor pulmonale. Oxygen therapy may reverse the component of pulmonary hypertension that is mediated by the presence of hypoxemia and may help to improve exercise performance and quality of life. Most of the data used to justify long-term oxygen therapy to improve survival in diffuse interstitial lung disease have been extrapolated from results obtained from long-term oxygen use in patients with emphysema and may not be relevant to patients with interstitial pulmonary fibrosis or usual interstitial pneumonia. Two studies, however, have assessed the acute effects of oxygen therapy in patients with a variety of interstitial lung diseases with moderate severity.76,77 The patients in these 2 studies all desaturated on exercise, and it is likely that oxygen therapy probably has a more significant effect in these individuals.
On the other hand, oxygen therapy may have little benefit or be harmful in patients who have idiopathic pulmonary fibrosis. Idiopathic pulmonary fibrosis in part may be due to oxygen mediated injury and oxygen therapy may increase tissue concentrations of toxic oxygen radicals. In a retrospective study of 487 Idiopathic pulmonary fibrosis patients, conducted by Douglas and colleagues, univariate analysis showed that oxygen use significantly shortened survival.78 However, when multivariate analysis was performed that was adjusted for age, sex, and disease severity, oxygen therapy was not found to be independently associated with a shortened survival. Their conclusion was that oxygen therapy was a marker of disease severity rather than being a toxic therapy that led to a shortened survival.
At present there are no prospective, randomized, and controlled trials that have examined the effect of short or long-term oxygen therapy on functional outcomes and survival in patients with idiopathic pulmonary fibrosis. However, it has been reported that oxygen therapy in this patient group is likely to improve other symptoms and overall quality of life, and has been used during rehabilitative exercise training to increase endurance.79,80 Currently oxygen therapy is considered to be a critical component of the management of idiopathic pulmonary fibrosis patients, but future data examining its use are required.
Congestive Heart Failure
CHF is the fastest growing cardiovascular disorder in the United States. It is estimated that CHF afflicts about 10 per 1,000 people who are ≥ 65 years of age. Six million people in the United States suffer from heart failure, and approximately 670,000 new cases are diagnosed annually.81 CHF is the most common cause for hospitalization in the elderly, and in the Medicare patient population heart failure is the most common reason for hospital admission and readmission within 30 days of hospital discharge. Medicare currently supports the use of long-term oxygen therapy and hypoxia-related conditions that may be expected to improve with oxygen therapy such as CHF or pulmonary hypertension. However, data from large randomized controlled trials that demonstrate the benefits of long-term oxygen therapy on acute or long-term outcome such as functional status or survival are also lacking in patients with CHF.
There is more robust data regarding the effects of oxygen therapy on sleep-disordered breathing that accompanies CHF. Cheyne-Stokes respiration is an abnormal respiratory pattern that is associated with CHF, and the crescendo-decrescendo alteration in tidal volume associated with periods of apnea or hypoxemia are thought to be compensatory mechanisms for fluctuating changes in the partial pressures of oxygen and carbon dioxide that result from reduced cardiac output in patients with advanced CHF. Hypoxemia clearly plays a central role in the initiation of Cheyne-Stokes respiration, and oxygen therapy may have a beneficial impact on decreasing its presence. In a randomized controlled study, Sasayama showed that nocturnal oxygen therapy in stable CHF patients with Cheyne-Stokes respiration and nocturnal hypoxemia may improve sleep disorder breathing, left ventricular function, and quality of life. Nocturnal oxygen therapy in CHF patients diagnosed with central sleep-disordered breathing and hypoxemia has also been shown to improve the apnea-hypoxemia index.82 Similar results have been reported in stable CHF patients with evidence of periodic breathing and hypoxemia that showed that nocturnal oxygen therapy improved periodic breathing and clinically important oxygen desaturation.83 Krachman and colleagues, in a prospective randomized controlled trial, compared nocturnal oxygen therapy with CPAP in patients with CHF and Cheyne-Stokes respiration, and concluded that nocturnal oxygen therapy and CPAP were equally effective in decreasing the apnea-hypopnea index in CHF patients with Cheyne-Stokes respiration.84
An increase in exercise tolerance and decrease in dyspnea have been noted when some patients with heart failure breathe enriched oxygen.85,86 However, these studies involved small number of subjects and had substantial variation in the methods used to perform exercise and the levels of oxygenation utilized, and most patients did not have hypoxemia at rest or during exercise without supplemental oxygen. Future investigation is required to determine the efficacy of oxygen therapy in patients with CHF.
Shortness of breath is common in patients with advanced cancer and may affect 50–70% of patients. Oxygen therapy to relieve breathlessness toward the end of life in patients with advanced malignancy is supported by consensus guidelines, and it is a common clinical practice.25 In a recent survey of over 600 palliative care specialist and pulmonary physicians, approximately 50% reported that palliative oxygen therapy is beneficial, and 65% reported that the most common reason for prescribing oxygen therapy was refractory dyspnea.87 Despite this clinical practice, however, data demonstrating that supplemental oxygen is effective in palliating dyspnea in patients with advanced malignancy are lacking. In a recent meta-analysis of 134 patients from 3 studies, oxygen therapy failed to improve dyspnea in mildly or non-hypoxemic cancer patients.88 Similarly, in a recent prospective randomized controlled trial of nasal air versus nasal oxygen in patients with life-threatening disease who were receiving palliative care for dyspnea (malignancy comprised about 15% of the population), nasal oxygen provided no benefit, compared to compressed air.25,89
At the present time it appears that oxygen therapy benefits patients with advanced malignancy who demonstrate hypoxemia at rest or with exertion or who exhibit a reduction in breathlessness in response to oxygen therapy. Future data are needed to better characterize the patient population with advanced cancer that may benefit from oxygen therapy.
Unresolved Issues Regarding Long-Term Oxygen Therapy
Despite its clinical use for over 40 years, many questions remain regarding the efficacy and appropriate indications for the use of long-term oxygen therapy in patients with advanced COPD, as well as in other diseases. Table 2 shows that Medicare approval for long-term oxygen therapy is based partially on expert opinion as well as evidence from positive randomized controlled trials.32,90 A recent National Heart, Lung, and Blood Institute workshop on long-term oxygen therapy in COPD highlighted the arbitrary nature of current clinical guidelines for oxygen use, the limited size and age of existing studies, limited knowledge regarding the duration and timing of oxygen use, and limitations in our knowledge regarding the optimal types of oxygen delivery devices (Table 3).32 Moreover, it is common knowledge that COPD patients frequently do not use their portable oxygen systems as prescribed by their physicians. In a recent report by Arnold et al, patients stated that they rarely received instructions from their physicians on how to use supplemental oxygen, they were unclear of the benefits of using oxygen, they were afraid that their oxygen device would run out of oxygen while using it, that the devices they were given for ambulatory oxygen were too heavy for them, and that the lack of a support person while using oxygen was an important obstacle to why they failed to use ambulatory oxygen therapy.91 Casaburi and colleagues further demonstrated that patients use their oxygen devices less than prescribed, even when provided with lightweight ambulatory portable units.92
In an attempt to determine whether supplemental oxygen is beneficial in patients with mild to moderate hypoxemia at rest or normoxic individuals at rest with mild to moderate exertional desaturation, the National Heart, Lung, and Blood Institute and Centers for Medicare and Medicaid Services have jointly enacted the Long-Term Oxygen Treatment Trial (LOTT).93 LOTT seeks to determine if supplemental oxygen increases the time to a composite outcome of all-cause mortality or all-cause hospitalization in COPD patients with mild to moderate degrees of resting hypoxemia (SpO2 89–93%) or resting SpO2 > 93% and moderate oxygen desaturation during exertion). Primary outcomes of LOTT include death or hospitalization, and secondary outcomes include time to death, time to hospitalization, and both disease specific and general effects of changes in quality of life. The goal is to recruit 737 eligible patients who will be followed for at least a year; over 500 patients have been recruited by this date. Hopefully, LOTT will provide answers regarding the efficacy of long-term oxygen therapy in patients with moderate resting hypoxemia, the efficacy of long-term oxygen therapy in those normoxic at rest but who desaturate during activity, the optimal methods for enhancing adherence to long-term oxygen therapy, and the optimal timing and dosing of oxygen supplementation.
Long-term oxygen therapy is an important disease-modifying treatment that provides significant improvements in physiologic and functional outcomes as well as survival in patients with advanced COPD who have severe hypoxemia at rest. Future investigation is needed to determine the effects of supplemental oxygen in patients with less severe forms of the disease, who have mild to moderate hypoxemia at rest or with exertion, as well as in other lung diseases that are characterized by hypoxemia to determine the efficacy of long-term oxygen therapy on survival and/or functional outcomes.
Being a former ARDS Network coordinator, I just want to say that I feel your pain about the problems of enrolling and randomizing patients into a clinical trial. I'm intrigued by how O2 toxicity relates to low-flow O2 therapy. I deal with this a little bit in my paper. Just finding an increased expression of pro-inflammatory mediators doesn't necessarily indicate something pathologically serious is going on. Did those studies measure anti-inflammatory mediators at the same time? If you have an inflammatory response from radical O2 species, you're basically going to have some sort of a counter-response, and it would be important to know that. Also, how important is the rise in pro-inflammatory mediators in disease progression? As a lover of strong coffee and good wine, an analogy comes to mind. Studies of coffee and alcohol have found that they produce trauma to your stomach lining, but it doesn't necessarily mean anything serious, at least in the short term. Does anybody have information on low-flow O2 toxicity?
I wanted to put the counter-argument out there. If you look at any chemokine or pro-inflammatory mediator, there's no global comprehensive description of all types of inflammatory and anti-inflammatory mediators. I could show you slides showing the anti-inflammatory effects of O2 therapy on decreasing neutrophil expression or inflammatory mediators that are found locally or systemically, but there is an imbalance in the literature as to whether it's pro- or anti-inflammatory.
Julio Turrens, who's one of the leading experts on mitochondrial damage, basically says that, under normobaric situations—unless it's the lungs we're talking about—it's really not an issue for other organs.1
It's difficult to take out the single-organ perspective of lung function without a concomitant coupling with the rest of the cardiovascular system. In terms of the metabolics and how these people respond, I would think you would be able to sort some of this out. Perhaps people who don't respond don't have an appropriate increase in cardiac output to activity, so you can't compensate for that just by raising arterial O2 content with supplemental O2. Or, in terms of afterload, reducing more people who don't respond so maybe they become responders. How do you differentiate between responders and non-responders, and can you make non-responders into potential responders if you're not going to take into account things like metabolics?
When we were designing the LOTT2 we discussed how to characterize the patients and whether they show predominately pulmonary benefit, or should we just forget about characterizing that? With what you mention, we have a patient group here who is older, frail, and with substantial comorbidities, including heart failure, cardiovascular disease, diabetes, or obesity, which all worsen their dyspnea and contribute to a linkage of pulmonary and cardiac dysfunction. After a laborious year and a half of designing the trial, we decided that if a patient comes into an emergency department or a hospital and we give them O2 and it improves their cardiac output, their cerebrovascular function, and it has an all-cause benefit on decreasing why they got admitted to the hospital or why they die, we don't exactly care if it improves the lung or the heart or red blood cell mass or decreases risk of stroke: we're looking for global patient benefit. It would be so exhaustive to characterize and then sub-sub-phenotype the patients by their physiologic criteria, where there's massive overlap in these patients. We decided it wasn't practical to do that.
But wouldn't you agree that it's eventually going to need to be done?
I think it probably will need to be done if it shows a benefit or an overwhelming negative and there's heterogeneity in the response. But overall I think that if it shows a benefit for this patient group—if it improves survival or hospitalization—I think that will carry the day.
In regard to the study that discussed the various subjective reasons why home oxygen patients didn't want to use their prescribed ambulatory O2 system, I believe the study was conducted in the United Kingdom, so these results might not translate globally. In the United Kingdom's home oxygen program, they use E cylinders almost exclusively as the ambulatory O2 system. The United Kingdom's health care system is very different from that of the United States, so those results may represent only the United Kingdom.
In regard to Rich Casaburi and colleagues' recent work3 comparing ambulation and activity in patients using E cylinders versus small, lightweight cylinders, the optimist in me saw about a 40% increase in ambulation in the small-cylinder group. In regard to the belief that the O2 device is the root of the ambulation problem, and if we give people the ideal portable O2 system they'll be out gallivanting around for 9 hours a day, I'm not quite sure we understand what this typically elderly patient population actually does in a given day or week. Are there any published comparisons to a control group of COPD patients who are not O2 dependent that measure and compare their activities of daily living and the amount of time they spend away from home versus those who are on O2? In my experience, even with the most active, adherent O2 patients you rarely see them using their ambulatory O2 for more than 2 hours a day.
In that study3 the patients are more severe than in LOTT: 75% of the group that was normoxic at rest desaturated. I think we'll probably get the answer from that trial, because those patients were less impaired than the COPD patients in LOTT.
How are you going to control the variability of the O2 products? I heard the patients are going to use their own O2 systems, and those systems are highly variable. And how do you make sure they're oxygenated at all activity levels?
The patients come in on a regular basis and have their O2 saturation measured at rest and during a 6-minute walk test, and they're titrated to maintain SpO2 at 90% at rest and exertion. There is a uniform sort of prescription that patients get for O2 therapy: a stationary and a lightweight ambulatory device, which is in the policy that goes out to all the vendors involved with this, and they know which are LOTT patients. It's coded and paid for by Medicare. So it's standardized in that way.
What is the pre-randomization treatment of these patients? How many times are you checking the O2 saturation? Is there a stabilization or pulmonary rehabilitation period before they get in, or is it just usual care before they're randomized?
It's standard community care: there's no phase-in period for pulmonary rehabilitation. The GOLD [Global Initiative for Chronic Obstructive Lung Disease] guidelines4 are used for optimizing medical care, and there are checklists on what medications the patients should be on.
And is it resting hypoxemia on one measurement or two?
We're using a quantitative system where we're measuring it for 5 minutes at rest, then there's a one minute period in which the desaturation has to occur.
During the designing of the trial were there any discussions that might provide insight into potential political or payer implications if it's a negative study?
It's not clear what the implications would be. We were told from the very beginning that Medicare has been involved with the meetings, and, depending on what we showed—and we were very careful with what we used for measures to describe the eligible patient population—that they were measures that could be used by anybody. If the trial is negative in harm, we don't know what changes in Medicare guidelines would occur.
Akin to sleep apnea with CPAP, where providers submit electronic data cards to verify use and maintain payment, has their been any discussion regarding installing monitoring cards into O2 ambulatory equipment to verify that patients actually use their equipment and thereby justify payment?
We designed a detailed adherence sub-study that was going to use the kind of device that was designed by the COPD Clinical Research Network, but that didn't come to be. Instead we're measuring patient usage by tanks delivered and self-reported days of use, but we don't have objective monitoring for each patient.
- Correspondence: Gerard J Criner MD, Division of Pulmonary and Critical Care Medicine, 845 Parkinson Pavilion, Temple University Hospital, 3401 N Broad Street, Philadelphia PA 19140. E-mail: .
Dr Criner presented a version of this paper at the 50th Respiratory Care Journal Conference, “Oxygen,” held April 13–14, 2012, in San Francisco, California.
The author has disclosed no conflicts of interest.
↵* Joseph S Lewarski RRT FAARC, Invacare, Elyria, Ohio.
- Copyright © 2013 by Daedalus Enterprises Inc.