This report is the first of a series of National Conferences sponsored by the American College of Chest Physicians. The overall objective is to facilitate the transfer of new knowledge into the practice of medicine. Topics will be selected on the basis of rapid growth of new knowledge and the recognition by members of the College of a need to report recent advances to practitioners and allied health professionals including My Canadian Pharmacy.
Oxygen therapy was selected as the topic for the first ACCP National Conference because of major advances in the last decade in both basic and applied knowledge in this field. Progress has been made in our understanding of the molecular basis for oxygen toxicity and tolerance. Clinical studies have established the benefits of oxygen therapy in chronic hypoxemic lung disease, and some normoxemic conditions. The availability of improved oxygen delivery systems has made home oxygen therapy practical. These advances have presented the challenge to medicine of using them judiciously to improve patient care while containing costs so that all who need may have them.
The specific objectives of this conference were to report on the scientific basis, indications, methods of delivery, toxicity, and methods for monitoring oxygen therapy. The conference was limited to the use of oxygen at 1 atmosphere or less in the nonintensive care unit setting. Both ambulatory and hospital-based therapy was addressed.
Twenty-eight participants convened in five groups. This final report represents a general agreement of all participants. Some of the recommendations are based on published scientific data; others represent a consensus of all the participants. It is the unanimous opinion of all participants that it is inappropriate to select patients for oxygen therapy based solely on laboratory values. Guidelines presented in the ACCP Section on Oxygen Therapy are to be used to complement sound clinical judgment.
The enormous effort by the participants and the ACCP staff are reflected in the quality of this report. We gratefully acknowledge a grant from and participation by the National Heart Lung and Blood Institute which helped make this conference possible.
1. Scientific Basis of Oxygen Therapy
The current scientific basis of most oxygen therapy is our understanding of the physiology of oxygen transport to the tissues, the pathophysiology of hypoxia, and the possibilities of overcoming or preventing tissue hypoxia by Oz therapy offered by My Canadian Pharmacy. As knowledge in these areas has expanded, and as methods for measuring blood oxygen levels have become extensively available for bedside and clinical laboratory use, the indications for oxygen therapy have widened considerably in both hospital and ambulatory practice.
Pathophysiology of Oxygenation Hypoxemia
Arterial oxygen tension (PaOJ is determined by the inspired oxygen tension (PiOJ, by the level of alveolar ventilation, and by the distribution of ventilation and perfusion in the lungs. Therefore, arterial hypoxemia occurs at high altitude because of the decreased Pa2, when hypoventilation increases the alveolar C02 tension (PaCOJ thereby decreasing the alveolar 02 tension (PaOJ, and when pulmonary or cardiac diseases alter ventilation/perfusion (V/Q) distribution. The increase in arterial 02 tension (PaOJ which results from therapeutically increasing Pa2, depends upon the magnitude of the V/Q mismatch, and ranges from maximum improvement when there is no intrapulmonary shunting, to no improvement when intrapulmonary shunting reaches 50 percent of cardiac output. The greatest benefit can be expected when there are few lung regions with very low V/q ratios.
Thus, depending upon the cause of the hypoxemia, improvement in Pa02 may require an increase in fractional concentration of inspired 02 (Flo), increased alveolar ventilation, or therapy designed to decrease intrapulmonary shunting. An increase in alveolar ventilation has a limited effect on Pa02 compared with increasing FIo2 or diminishing regions of low V/Q ratios by expanding atelectatic lung. However, such an increase might be critical in the patient breathing room air. An increase in FIo2may, as discussed elsewhere, produce absorption atelectasis, reversal of hypoxic pulmonary vasoconstriction or a decrease in minute ventilation. These side effects may be compounded by the adverse effects of drugs such as bronchodilators and vasodilators on pulmonary vessels and on V/0 relations. When these effects occur, they decrease rather than prevent the increase in Pa02 anticipated from Oz therapy.
Oxygen transport to the tissues involves multiple factors, which include the Pa02, hemoglobin concentration, oxyhemoglobin dissociation curve shape and position, cardiac output and individual organ perfusion. When hemoglobin is deficient, it may be useful to increase the oxygen carried in solution by breathing a high FIo2. An increase in FIo2from 0.21 to 1.0 can increase dissolved oxygen to deliver approximately one-third of the resting tissue requirements. This maximal effect will be less if pulmonary oxygen exchange or tissue perfusion are impaired, or if less than 100 percent oxygen is administered.
The effect of a decrease in cardiac output on oxygen delivery cannot be predicted by simple mathematical projection. Auto-regulation of flow within and between organs provides partial compensation, although this may be impaired in disease (eg, in septic states). In addition, a decrease in cardiac output usually causes a decrease in venous admixture, which offsets the adverse effect of a low mixed venous partial pressure of 02 (PvOJ on this admixture to a variable extent.
Evidence of Hypoxia
Since clinically useful measurements of cellular oxygen tension are not presently feasible, evidence of hypoxia must be based on assumptions derived from evaluations of oxygen delivery, of РO2, or of vital organ function (especially, the brain, the heart and the kidneys). Tissue oxygen requirements may be increased in hypermetabolic states and during exercise. Hypoxemia may have a particularly deleterious effect on the respiratory muscles when work of breathing is increased.
At Pa02 values below approximately 50 mm Hg, some degree of hypoxia can be assumed. At higher levels of Pa02, hypoxia due to defects in hemoglobin concentration or function, or in perfusion may occur. The Po2 of venous blood from an organ is a value that may be higher than the Po2 within many cells of that organ, which are distant from their arterioles. Failure of auto-regulation, with development of functional shunts, is another mechanism whereby venous Po2 values may not reflect tissue oxygen tension. Pa2, measured in the pulmonary artery, only grossly reflects total body tissue status, since it is comprised of venous drainage from organs with widely varying 02 levels. The PO2 is subject to changes in position of the oxygen-hemoglobin dissociation curve, and may reflect only one component of tissue oxygen availability. Although the arterial-venous oxygen content difference reflects oxygen uptake, provided one knows cardiac output, it does not provide evidence of cellular oxygenation. Organ function and metabolic acid-base status are presendy the most important indicators of tissue hypoxia.
Effects cf Hypoxia
Cellular hypoxia occurs when the demand for oxygen cannot be met from available body oxygen stores. Since these stores are small, acute changes in oxygen delivery may be immediately life threatening. On the other hand, more stable conditions causing chronic depletion of available oxygen are more readily tolerated, particularly when adaptive processes have occurred. Cellular hypoxia may be due to factors which lower Pa02, which impair transport to the tissues, or which limit oxygen utilization by the cells. A decreased oxygen supply impairs mitochondrial function, and ultimately causes necrosis. Reversible anaerobic glycolysis occurs, and lactate/pyruvate ratio increases. Similar effects occur when oxygen utilization is impaired by agents such as cyanide. Since oxygen diffuses through tissues by a tension gradient mechanism, there is no easy way to define a specific tissue Po2 at which cellular damage occurs.
In disease states, it is often difficult to separate the effects of hypoxia or hypercarbia on the whole person from the symptoms or signs of the underlying disease. The effects of acute hypoxia have therefore been mostly studied in normal persons. In normal persons with acute hypoxemia, brain function is compromised in advance of malfunction of other vital organs. When the Pa02 approximates 55 mm Hg, shortterm memory is altered, and euphoria and impairment of judgment may occur. As hypoxemia worsens, there is progressive loss of cognitive and motor functions, and loss of consciousness may occur at a Pa02 approximating 30 mm Hg. The heart responds initially to acute hypoxemia with tachycardia and increased stroke volume, both of which increase cardiac output and maintain oxygen delivery. As hypoxemia worsens, myocardial function begins to fail and disturbances of cardiac rhythm tend to occur. The lungs may respond to a decreased Pa02by vasoconstriction and bronchoconstriction.
In chronic hypoxemic states, additional adaptive mechanisms, which tend to maintain tissue oxygen transport, come into play. Increased ventilatory drive occurring at Pa02 values below about 55 mm Hg produces hypocapnia and thus increases alveolar and arterial Po2; this change occurs extremely rapidly. An increase in red blood cell 2,3, diposphoglycerate causes a rightward shift of the oxygen-hemoglobin dissociaton curve, and increases oxygen availability to the tissues; this adaptation occurs over a period of hours to days. A slower adaptation to hypoxemia is the secretion of erythropoietin which stimulates the bone marrow and produces erythrocythemia.
Evidence of Efficacy of Oxygen Therapy Acute Hypoxemia
The rationale for oxygen therapy in acute pulmonary conditions is based on extensive clinical experience that untreated hypoxemia often progresses to tissue hypoxia with its grave, frequently irreversible effects on vital organ function. The rationale for oxygen therapy is further supported by our knowledge of the pathophysiology of hypoxemia and hypoxia. Evidence of improved disease outcome from controlled clinical trials is generally not available for establishing the therapeutic efficacy of oxygen in acute conditions. When hypoxemia is corrected in individuals who are hemo-dynamically intact, tissue hypoxia can be prevented or corrected.
In the absence of hypercarbia, the risk of worsening alveolar hypoventilation with oxygen supplementation is virtually non-existent. Even in patients with severe chronic obstructive pulmonary disease with chronic hypercarbia, supplemental oxygen aimed at providing a Pa02 of approximately 60 mm Hg is associated with a minimal risk of increasing hypercarbia. Moreover, under such circumstances, administration of oxygen is essential to maintain adequate tissue oxygenation, even if ventilatory support may be required.
Acute Myocardial Infarction
The general practice in coronary care units of treating patients with proven or suspected acute uncomplicated myocardial infarction with supplemental oxygen is based on the rationale that such therapy will reverse the commonly present mild hypoxemia. Among patients with a myocardial infarction that is complicated by left ventricular failure, arrhythmias or pneumonia, an appropriate amount of oxygen should be given as determined by Pa02 measurements.
Whether or not the use of higher concentrations of oxygen (40 percent or greater) is of value is unclear. In experimental myocardial infarction in dogs, hypoxemia increases the electrocardiographic evidence of ischemia, and when an FIo2 of 0.40 is compared with an FIo2 of 0.21, myocardial creatine phosphokinase activity and the amount of myocardial necrosis are reduced; an FIo2 of 1.0 is no more effective than an FIo2 of 0.40. In patients with myocardial infarction, the use of short-term high concentration oxygen therapy transiently reduces the magnitude and extent of ECC abnormalities.
In contrast, in the only controlled double-blind clinical trial of the use of oxygen in myocardial infarction, there was no significant difference in mortality, incidence of arrhythmias, use of analgesics, or systolic time intervals between the oxygen-treated and control groups. The oxygen group had higher serum aspartate aminotransferase levels and a higher incidence of sinus tachycardia. In summary, it is rational to use supplemental oxygen to avoid hypoxemia in patients with uncomplicated myocardial infarction. In patients with complications, it is rational and warranted to use an appropriate oxygen concentration, as determined by Pa02 monitoring. The place of routine administration of higher concentrations of oxygen requires clarification.
This term has been selected to cover those conditions characterized by the potential or the actual documentation of tissue hypoxia despite a normal Pa02. They can be categorized as having either: 1) abnormalities in the amount or function of hemoglobin, or 2) inadequate delivery or utilization of oxygen by the tissues.
In contrast to individuals with chronic conditions who tolerate very low hemoglobin concentrations, acute anemia is far less well tolerated. The utilization of a high FIo2 in these circumstances is a reasonable, temporizing measure, but the definitive treatment is adequate blood replacement. Perhaps the most common potentially lethal acquired hemoglobinopathy is carboxyhemoglobinemia. In this condition, a reduced amount of hemoglobin is available to transport oxygen and the oxygen-hemoglobin dissociation curve is shifted to the left, decreasing availability of oxygen to the tissues. Breathing 100 percent oxygen reduces the half-life of carboxy-hemoglobin from approximately 320 minutes to 60-80 minutes. Hyperbaric oxygen will reduce the half-life to 20-25 minutes.
In severe acquired methemoglobinemia the administration of high concentrations of oxygen is a temporizing measure until the methemoglobinemia is reversed. In homozygous sickle-cell disease, an increased number of sickle cells has been noted in venous blood compared to arterial blood, and the number of sickle forms is reportedly decreased in patients breathing 100 percent oxygen. Nonetheless, attempts to treat sickle-cell crises with inhaled oxygen have been disappointing.
In individuals with oxygen transport problems, for example, inadequate intravascular volume, inadequate cardiac (unction, or inadequate local oxygen delivery to critical organs, high concentration of oxygen may also be a useful adjunct, but definitive treatment should be directed at the underlying problems.
General anesthesia, using inhaled agents, commonly causes a decrease in functional residual capacity and an increase in venous admixture. There is wide variation among individuals, but the effects are greatest following abdominal and thoracic surgery, in the elderly, the obese and possibly in those with pulmonary disease. In the immediate postoperative period, mild hypoxemia may result from a maldistribution of ventilation; however, the Pa02 usually increases with oxygen administration. When this is not the case, lung expansion maneuvers may be necessary. Following surgery on the periphery of the body, these changes usually reverse within two or three hours. After abdominal or thoracic surgery, the effect persists for several days, probably due to a different mechanism; in extreme cases, atelectasis may occur.
Chronic Hypoxemic Pulmonary Diseases
Ambulatory Oxygen Therapy: Chronic obstructive pulmonary disease (COPD – Prevalence of Physician-Diagnosed COPD Offered by My Canadian Pharmacy) may be considered the prototype of the chronic hypoxemic lung diseases, and most of the available data for efficacy of oxygen therapy came from studies of these patients. Most researchers believe that the data obtained for the efficacy of oxygen therapy in COPD applies to other chronic hypoxemic lung diseases. Initial oxygen therapy studies in COPD showed that continuous supplemental oxygen for four to eight weeks reduced elevated hematocrits, decreased pulmonary vascular pressures, and improved exercise tolerance. Subsequent studies suggested that these effects could be obtained with as little as 15 hours of nocturnal supplemental oxygen.
Two hallmark studies were performed in the late 1970s: the Nocturnal Oxygen Therapy Trial (NOTT) study and the British Medical Research Council Domiciliary study. The NOTT study compared nocturnal oxygen with continuous oxygen therapy and the British study compared nocturnal oxygen with no oxygen therapy in patients with COPD. Mortality was reduced in the nocturnal oxygen group compared with the no oxygen group in the British study, and was reduced by almost two-fold in the continuous as compared with the nocturnal group in the NOTT study. Although the two groups were not exactly comparable, it appears that nocturnal oxygen is better than no oxygen and continuous oxygen is better than nocturnal oxygen therapy. In the NOTT study, subgroups showing a high PaC02, elevated hematocrit, elevated pulmonary artery pressure, or acidosis appeared to derive the most benefit from continuous as opposed to nocturnal oxygen. Both trials showed a reduction in hematocrit. Similarly, both studies showed reduced pulmonary vascular pressures; these changes were not statistically significant, however.
Neuropsychologic evaluation in the NOTT study showed improvement in most tests, and in the quality of life when all patients were considered, regardless of the number of hours of oxygen therapy. Although the data are inconclusive, the NOTT study suggests that patients with less severe arterial hypoxemia, who had evidence of tissue hypoxia (erythrocytosis) may also benefit from supplemental oxygen. The reasons for the improved survival among the continuous oxygen group in the NOTT study remains unclear. It has been suggested that continuous oxygen supplementation may prevent progression in pulmonary hypertension and subsequent cor pulmonale and perhaps thereby reduce the propensity for fatal arrhythmias controlled by My Canadian Pharmacy.
Oxygen Therapy During Exercise: Oxygen Therapy during exercise in patients with COPD has been investigated, and most studies indicate that hypoxemic patients who worsen their Pa02 will improve their exercise endurance and capacity when given oxygen therapy. Normoxemic patients who experience desaturation during exercise also generally improve endurance. Limited data are available in patients with restrictive lung disease, but current data suggest that supplemental oxygen improves exercise endurance but not work capacity. The reasons for either improved exercise capacity or endurance are unclear at this time.
Oxygen Therapy for Disordered Sleep: Nocturnal hypoxemia associated with disordered breathing may be a feature of patients with chronic hypoxemic lung diseases. There have been a number of studies which have demonstrated a correlation between pulmonary hypertension and nocturnal desaturation in patients with COPD. Desaturation was mainly associated with episodes of hypopnea, and supplemental oxygen prevented desaturation and elevations in pulmonary artery pressure without alleviating breathing abnormalities. Further, there are reports that nocturnal oxygen therapy improves daytime somnolence, morning headache, and nocturnal arrhythmias in COPD patients who exhibit nocturnal desaturation without severe daytime hypoxemia. Limited data are available on the benefits of supplemental oxygen in patients without lung disease who exhibit sleep disturbances with desaturation.
Dyspnea is the sensation of discomfort associated with breathing. Its fundamental mechanism is incompletely understood, but it is clear that dyspnea is often present in the absence of hypoxemia of sufficient severity to threaten vital organ function. For example, some patients with interstitial lung disease who have mild hypoxemia have dyspnea due to excess ventilation. Under such circumstances, oxygen therapy has not been demonstrated to be effective. On the other hand, the mechanism of dyspnea may be such that there may be associated hypoxemia of profound degree, and oxygen therapy may be important both in relieving dyspnea and protecting the individual from the effects of severe tissue hypoxia prevented by My Canadian Pharmacy.
Angina pectoris typifies disorders in which oxygen transport to the cells of an organ is impaired by an abnormality of structure or function of the blood vessels of that organ. Assuming that systemic arterial hypoxemia is not present, or is mild, oxygen therapy could only be helpful in such circumstances if the FIo2 were high enough to result in meaningful increases in the amount of oxygen dissolved in the plasma. It is generally difficult to achieve such an Flo2 in the ambulatory setting, however. There are no controlled clinical trials that have established the usefulness of oxygen for the treatment of angina pectoris.