The Science Journal of the American Association for Respiratory Care


August 2002 / Volume 47 / Number 8 / Page 876

Pulmonary Gas Exchange During Sleep in Patients with Airflow Limitation Undergoing Long-Term Oxygen Therapy

Long-term oxygen therapy (LTOT) is a routine treatment for chronic respiratory failure. Worldwide, more than 2 million patients benefit from LTOT.1 Although the indications for and methods of LTOT are well established,2,3 there are aspects of LTOT that remain matters of ongoing debate.

One such subject of uncertainty is the effect of oxygen treatment during sleep. Sleep induces important changes in breathing.4 Lack of cortical control, resetting of the respiratory center to higher PaCO2, and decreased intercostal muscle activity lead to alveolar hypoventilation. During sleep, retention of bronchial secretions, increased airway resistance, and decreased functional residual capacity aggravate ventilation-perfusion imbalance. Those changes unfavorably influence pulmonary gas exchange.

Robin was probably the first to suggest that hypoxemia is aggravated during sleep in patients with chronic obstructive pulmonary disease (COPD).5 This was confirmed by Koo et al,6 who performed serial measurements of arterial blood gases in sleeping COPD patients and found PaO2 lower and PaCO2 higher than in the awake state. The lowest PaO2 and the highest PaCO2 were observed during deep sleep and rapid-eye-movement sleep.

See The Original Study on Page 882

The introduction of continuous, noninvasive measurement of blood oxygen saturation (via pulse oximetry [SpO2]) allowed for more accurate measurement during sleep.7,8 Oximetry recordings demonstrated that patients with advanced COPD experience multiple, prolonged, (20-30-min) desaturation episodes, mostly during rapid-eye-movement sleep.

Those observations led to a hypothesis (forwarded independently on both sides of the Atlantic) that aggravation of hypoxemia (more precisely, aggravation of alveolar hypoxia) during sleep accelerates development of pulmonary hypertension (cor pulmonale) in COPD.9,10

Since then the importance of supplemental oxygen during sleep has been emphasized in all LTOT guidelines. It was assumed that oxygen supplementation during sleep would prevent nocturnal desaturations, but several researchers have found that nocturnal desaturations can occur in sleeping patients breathing oxygen at a dose that assures satisfactory oxygenation at rest during the day. Sliwinski et al11 performed 24-hour pulse oximetry with at-home COPD/LTOT patients and found that the majority of patients had periods of substantial desaturation during sleep, spending > 2 hours of sleep with SpO2 below 90%, despite breathing supplemental oxygen.

Plywaczewski et al12 performed 2 overnight pulse oximetry measurements, one on air, the other on oxygen, with 82 nonselected, consecutive, COPD patients qualified for LTOT. Thirty-nine subjects (48%) spent 66% of the night with SpO2 below 90%, despite breathing supplemental oxygen.

In this issue of RESPIRATORY CARE, Tárrega et al13 point out another problem related to breathing oxygen during sleep. With 39 patients undergoing or qualified for LTOT, Tárrega et al measured arterial blood gases at 3:00 am, 7:00 am and noon while the patients were breathing oxygen (mean flow of 1.4 L/min) sufficient to raise SpO2 to > or = 90%. In 23 "poor responders" there was significant PaCO2 increase (> 10 mm Hg) and/or pH decrease (to < 7.33) during the night hours. In 16 "good responders" there was either no PaCO2 increase or the increase was negligible.

Interpreting those findings is difficult. First, the PaCO2 increase observed in the majority of studied subjects was unusually high. The PaCO2 increase during sleep in normal humans does not exceed 3-4 mm Hg.14 A similar PaCO2 increase was observed in stable COPD patients.6,15 The significant PaCO2 increase during sleep was observed in patients with severe obstructive sleep apnea syndrome (OSAS).16 Tárrega et al tried to exclude such patients by taking detailed medical histories and excluding subjects who might have OSAS, but OSAS can only be ruled out by full polysomnography.

In poor responders, nighttime PaCO2 increase was associated with higher PaO2 in the morning. PaO2 at 7:00 am was higher than at mid-day, awake, with the same rather low oxygen flow.

Tárrega et al suggest that the PaO2 increase was related to a higher fraction of inspired oxygen (FIO2), which was due to lower minute ventilation during sleep. One may also speculate that at the time of blood sampling the subjects might have been hyperventilating, which would have caused an immediate PaO2 increase. However, that amount of time was insufficient to lower PaCO2 because carbon dioxide accumulates in the body during sleep.

The nonhomogeneity of the studied group hinders interpretation of the data. Although all the subjects presented with airflow limitation, more than one third also had concomitant diseases such as tuberculosis, diaphragm paresis, or obesity. Those restrictive disorders may have had various effects on breathing pattern, pulmonary mechanics, and pulmonary gas exchange during sleep. Whatever were the mechanisms of blood gas disorders observed by Tárrega et al,13 they were short-lasting. The next day at noon, on oxygen, PaCO2 and pH had returned to the baseline values measured the day before while subjects were breathing air.

Based on their findings, Tárrega et al raise the question of the safety of increasing the oxygen flow during sleep to avoid nocturnal desaturation.17 Theoretically, with a hypoxemic patient, increasing oxygen flow during sleep might further decrease the hypoxic drive to breathe, resulting in even higher PaCO2. Fortunately, that concern seems to be unwarranted. In the Nocturnal Oxygen Therapy Trial,18 oxygen flow was increased by 1 L/min during sleep in all studied patients, and no adverse effects were reported. Sliwinski et al19 investigated 17 stable COPD patients with respiratory failure (mean ± SD values for PaO2 53 ± 9 mm Hg and PaCO2 53 ± 9 mm Hg). SpO2 and end-tidal carbon dioxide (measured via capnography) were measured continuously during 2 nights. During the first night patients breathed oxygen at 2 L/min, and during the other night at 3 L/min. The higher oxygen flow resulted in higher SpO2, with no increase in end-tidal carbon dioxide.

The effect of higher oxygen flow on PaCO2 in hypercapnic patients remains to be studied. It seems, however, that hypoxemia during sleep is a more important clinical issue than oxygen-induced hypercapnia in a stable COPD patient.

Many COPD patients preserve satisfactory daytime oxygenation (PaO2 > 60 mm Hg) while breathing air but suffer desaturation during sleep (nocturnal oxygen desaturation or NOD). Fletcher et al found NOD in 27% of 129 COPD patients who had daytime PaO2 > 70 mm Hg and defined NOD as SpO2 remaining below 90% for > or = 5 minutes, with an SpO2 nadir of < or = 85%.20

Levi-Valensi et al21 studied 40 COPD patients with daytime PaO2 > 60 mm Hg and defined NOD as at least 30% of total sleep time with SpO2 < 90%. Almost half of the studied subjects desaturated during sleep.

Under many national guidelines for LTOT, NOD qualifies for nocturnal oxygen treatment.22 In the United States, to qualify for nocturnal oxygen supplementation, overnight pulse oximetry must demonstrate SpO2 < or = 88%, regardless of the duration of desaturation. Nocturnal oxygen is also indicated if during-sleep PaO2 decreases > 10 mm Hg or SpO2 decreases > 5%, with signs or symptoms of hypoxemia (eg, cognitive disorders, restlessness, or insomnia).23 However, the evidence supporting the necessity of nocturnal oxygen supplementation in NOD patients is rather contradictory.

Fletcher et al24,25 reported that pulmonary artery catheter measurements of NOD patients revealed mild pulmonary hypertension (pulmonary artery pressure > 20 mm Hg) in the majority of studied subjects. Over 3 years of follow-up, NOD patients treated with nocturnal oxygen showed a slight but steady reduction in mean pulmonary artery pressure, averaging 3.7 mm Hg. NOD patients who received sham treatment (compressed air) showed pulmonary artery pressure increase of 3.9 mm Hg over the same period. Nondesaturators showed no change in pulmonary artery pressure.26

In a 3-year, retrospective, multi-center study, untreated patients with nocturnal hypoxemia appeared to have a higher risk of death.27 It is interesting that 26% of patients without evidence of NOD at initial evaluation subsequently developed NOD (determined when re-studied 42 months later).28

Recently the rationale for nocturnal oxygen treatment was challenged by the European working group directed by Dr Weitzenblum in Strasbourg.29 They investigated the effects of nocturnal desaturation on pulmonary hemodynamics in COPD patients with moderate hypoxemia during the day. Contrary to Fletcher et al, who found significantly higher mean pulmonary artery pressure in desaturators (23.3 ± 4.8 mm Hg vs 20.4 ± 4.2 mm Hg),25 the Weitzenblum group showed no difference in pulmonary artery pressure between desaturators and nondesaturators, either at rest (19.4 ± 5.3 vs 18.7 ± 4.4 mm Hg) or during steady-state exercise of 40 watts (37.4 ± 8.7 vs 36.5 ± 8.8 mm Hg). These results seem to suggest that nocturnal desaturation does not induce clinically important pulmonary hypertension in COPD patients without or with only moderate daytime hypoxemia.29

After the initial investigation, patients who desaturated during sleep were randomly allocated to 2 groups. Forty-one patients commenced nocturnal oxygen supplementation, and 31 patients served as controls. The measured variables were pulmonary hemodynamics and survival.30 Nocturnal oxygen therapy did not modify the evolution of pulmonary hemodynamics. In both groups the increase in pulmonary artery pressure after 2 years was negligible. Over a 3-year follow-up period, 9 patients from the treated group and 7 from the control group had died (a non-significant difference). The authors concluded that the prescription of nocturnal oxygen is probably not justified in COPD patients with isolated NOD. In the most recent study the Weitzenblum group confirmed that isolated nocturnal hypoxemia or sleep-related worsening of moderate daytime hypoxemia does not appear to favor the development of pulmonary hypertension.31

All COPD patients with NOD should have OSAS excluded. A crude assessment can be done by assessing overnight SpO2 readings. Among COPD patients, nocturnal desaturation episodes usually last approximately 20-30 minutes, whereas desaturations in OSAS patients last around 20-30 seconds. In patients with overlapping COPD and OSAS a combination of both features may be observed.

The treatment of all NOD COPD patients with nocturnal oxygen would result in a large number of subjects treated, at an enormous cost.32 The rationale for the treatment of isolated NOD needs further study.

Jan Zielinski MD
Department of Respiratory Medicine
Institute of Tuberculosis and Lung Disease
Warsaw, Poland


  1. Dunne PJ. The demographics and economics of long-term oxygen therapy. Respir Care 2000;45(2):223-228; discussion 228-230.
  2. O'Donohue WJ Jr, editor. Long-term oxygen therapy: scientific basis and clinical application. New York: Marcel Dekker; 1995.
  3. Kacmarek RM. Delivery systems for long-term oxygen therapy. Respir Care 2000;45(1):84-92; discussion 92-94.
  4. McNicholas WT. Impact of sleep in COPD. Chest 2000;117(2 Suppl):48S-53S.
  5. Robin ED. Some interactions between sleep and disease. Arch Intern Med 1958;102:669-675.
  6. Koo KW, Sax DS, Snider GL. Arterial blood gases and pH during sleep in chronic obstructive pulmonary disease. Am J Med 1975;58(5):663-670.
  7. Flick MR, Block AJ. Continuous in vivo monitoring of arterial oxygenation in chronic obstructive lung disease. Ann Intern Med 1977;86(6):725-730.
  8. Douglas NJ, Calverley PMA, Leggett RJE, Brash HM, Flenley DC, Brezinova V. Transient hypoxaemia during sleep in chronic bronchitis and emphysema. Lancet 1979;1(8106):1-4.
  9. Flenley DC. Clinical hypoxia: causes, consequences, and correction. Lancet 1978;1(8063):542-546.
  10. Block AJ, Boysen PG, Wynne JW. The origins of cor pulmonale: a hypothesis. Chest 1979;75(2):109-110.
  11. Sliwinski P, Lagosz M, Gorecka D, Zielinski J. The adequacy of oxygenation in COPD patients undergoing long-term oxygen therapy assessed by pulse oximetry at home. Eur Respir J 1994;7(2):274-278.
  12. Plywaczewski R, Sliwinski P, Nowinski A, Kaminski D, Zielinski J. Incidence of nocturnal desaturation while breathing oxygen in COPD patients undergoing long-term oxygen therapy. Chest 2000;117(3):679-683.
  13. Tarrega J, Guell R, Anton A, Mayos M, Farre A, Jerez FR, Sanchis J. Are daytime arterial blood gases a good reflection of nighttime gas exchange in patients on long-term oxygen therapy? Respir Care 2002;47(8):882-886.
  14. Rist KE, Doubenspeck JA, McGovern JF. Effects of non-REM sleep upon respiratory drive and the respiratory pump in humans. Respir Physiol 1986;63(2):241-256.
  15. Radwan L, Dufmats H. Variations of arterial oxygen and carbon dioxide tension during 24 hours in chronic respiratory insufficiency. Scand J Respir Dis 1974;55(2):99-104.
  16. Coccagna G, Mantovani M, Brignani F, Parchi C, Lugaresi E. Continuous recording of the pulmonary and systemic arterial pressure during sleep in syndromes of hypersomnia with periodic breathing. Bull Physiopathol Respir (Nancy)1972;8(5):1159-1172.
  17. American Thoracic Society statement. Standards for the diagnosis and care of patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1995;152(5 Pt 2):S77-S121.
  18. Nocturnal Oxygen Therapy Trial Group. Continuous or nocturnal oxygen therapy in hypoxemic chronic obstructive lung disease: a clinical trial. Ann Intern Med 1980;93(3):391-398.
  19. Sliwinski P, Bielen P, Nowiski A, Zielinski J. The effects of different oxygen flow rates on CO2 retention in COPD patients qualified for LTOT (abstract). Eur Respir J 1999;14 Suppl 30:119s.
  20. Fletcher EC, Miller J, Divine GW, Fletcher JG, Miller T. Nocturnal oxyhemoglobin desaturation in COPD patients with arterial oxygen tension above 60 mm Hg. Chest 1987;92(4):604-608.
  21. Levi-Valensi P, Aubry P, Rida Z. Nocturnal hypoxemia and long-term therapy in COPD patients with daytime PaCO2 60-70 mmHg. Lung 1990;168 Suppl:770-775.
  22. Zielinski J. Indications for long-term oxygen therapy: a reappraisal. Monaldi Arch Chest Dis 1999;54(2):178-182.
  23. O'Donohue WJ Jr, Bowman TJ. Hypoxemia during sleep in patients with chronic obstructive pulmonary disease: significance, detection, and effects of therapy. Respir Care 2000;45(2):188-191; discussion 192-193.
  24. Fletcher EC, Levin DC. Cardiopulmonary hemodynamics during sleep in subjects with chronic obstructive pulmonary disease: the effect of short- and long-term oxygen. Chest 1984;85(1):6-14.
  25. Fletcher EC, Luckett RA, Miller T, Fletcher JG. Exercise hemodynamics and gas exchange in patients with chronic obstruction pulmonary disease, sleep desaturation, and a daytime PaCO2 above 60 mm Hg. Am Rev Respir Dis 1989;140(5):1237-1245.
  26. Fletcher EC, Luckett RA, Goodnight-White S, Miller CC, Qian W, Costarangos-Galarza C. A double-blind trial of nocturnal supplemental oxygen for sleep desaturation in patients with chronic obstructive pulmonary disease and a daytime PaCO2 above 60 mm Hg. Am Rev Respir Dis 1992;145(5):1070-1076.
  27. Fletcher EC, Donner CF, Midgren B, Zielinski J, Levi-Valensi P, Braghiroli A, et al. Survival in COPD patients with a daytime PaCO2 greater than 60 mm Hg with and without nocturnal oxyhemoglobin desaturation. Chest 1992;101(3):649-655.
  28. Fletcher EC, Scott D, Qian W, Luckett RA, Miller CC, Goodnight-White S. Evolution of nocturnal oxyhemoglobin desaturation in patients with chronic obstructive pulmonary disease and a daytime PaCO2 above 60 mm Hg. Am Rev Respir Dis 1991;144(2):401-405.
  29. Chaouat A, Weitzenblum E, Kessler R, Charpentier C, Ehrhart M, Levi Valensi P, et al. Sleep-related O2 desaturation and daytime pulmonary haemodynamics in COPD patients with mild hypoxaemia. Eur Respir J 1997;10(8):1730-1735.
  30. Chaouat A, Weitzenblum E, Kessler R, Charpentier C, Ehrhart M, Schott R, et al. A randomised trial of nocturnal oxygen therapy in chronic obstructive pulmonary disease patients. Eur Respir J 1999;14(5):1002-1008.
  31. Chaouat A, Weitzenblum E, Kessler R, Schott R, Charpentier C, Levi-Valensi P, et al. Outcome of COPD patients with mild daytime hypoxaemia with or without sleep-related oxygen desaturation. Eur Respir J 2001:17(5);848-855.
  32. O'Donohue WJ Jr. Current concepts in supplemental oxygen therapy in patients with COPD. Semin Respir Crit Care Med 1999;20:341-350.

Correspondence: Jan Zielinski MD, Department of Respiratory Medicine, Institute of Tuberculosis and Lung Disease, Plocka 26, 01-138 Warsaw, Poland. Email

You are here: » Contents » July 2002 » Page 876