Cardiovascular effects of breathing 95 percent oxygen in children with congenital heart disease

The hemodynamic effects of breathing 95% oxygen were evaluated in 26 children with congenital heart disease. Aortic, pulmonary arterial, right atrial, and pulmonary arterial wedge pressure, aortic and pulmonary artery oxygen saturation, and blood gas, cardiac index, and heart rate were measured in room air and after each patient had breathed 95 % oxygen for 10 (n = 26) and 20 (n = 5) minutes. Measurements were repeated with the patient again breathing room air for 10 (n = 11) and 20 (n = 6) minutes. After 10 minutes of 95% oxygen, arterial partial pressure of oxygen increased from 85 ± 13 to 420 ± 89 torr (p < 0.001). Aortic mean pressure increased from 80 ± 10 to 83 ± 10 mm Hg (p < 0.01), and systemic vascular resistance increased from 20 ± 7 to 26 ± 8 U (p < 0.001). The cardiac index decreased by 21 % from 3.96 ± 0.94 to 3.12 ± 0.74 liters/min/m² (p < 0.001) and the stroke index decreased by 11% (p < 0.001). A 23% decrease in oxygen consumption (p < 0.001) was observed, and oxygen transport decreased from 763 ± 179 to 600 ± 161 ml O₂/min/m² (p < 0.001). Cardiac index, stroke index, and systemic vascular resistance did not return to normal until 20 minutes after cessation of oxygen breathing. To determine whether reflex bradycardia is responsible for these oxygen-induced hemodynamic changes, heart rate was kept constant by atrial pacing in a second group of 5 patients. In these children, significant decreases in cardiac index, stroke index, and oxygen consumption, and increases in systemic vascular resistance also occurred with 95% oxygen. Thus, in children with acyanotic congenital heart disease, hyperoxia increases aortic pressure and systemic vascular resistance and decreases cardiac index, stroke index, oxygen consumption, and oxygen transport.     

Radionuclide angiocardiographic assessment of pulmonary vascular reactivity in patients with left to right shunt and pulmonary hypertension

Radionuclide angiocardiography was used to assess pulmonary vascular reactivity in eight patients (nine studies) with a large, relatively unrestrictive intracardiac defect and pulmonary arterial hypertension. Radionuclide angiocardiograms, using technetium-99m pertechnetate, were performed first with the patient breathing room air and then after 10 minutes of breathing a mixture containing 90 percent or more of oxygen. The pulmonary to systemic flow ratios obtained by gamma variate analysis of the radionuclide time-activity curves were compared with those calculated with the Fick principle at the time of cardiac catheterization. There was a good correlation between the two methods both in room air studies (r = 0.88) and in those obtained with 90 percent or more of oxygen (r = 0.94). All six studies (in five patients) with a reactive pulmonary vasculature (judged by a pulmonary vascular resistance at cardiac catheterization of less than 6 units/m2 with oxygen or after tolazoline) had a radionuclide pulmonary to systemic flow ratio of 3.0 or greater with oxygen. The three patients with a nonreactive pulmonary vasculature had a radionuclide pulmonary to systemic flow ratio of 2.3 or less with oxygen, a value that was unchanged from the room air value. These data suggest that radionuclide angiocardiography may be a useful, relatively noninvasive method of assessing pulmonary vascular reactivity in patients with a large, relatively unrestrictive intracardiac defect.     

Vascular and cardiac reactivity in pulmonary hypertension due to chronic obstructive lung disease: assessment with various oxygen concentrations.

The aim of the present work was to evaluate vasoreactivity in patients with pulmonary hypertension related to chronic obstructive lung disease. This was done by comparing haemodynamic data recorded while patients were breathing room air, and hypoxic and hyperoxic mixtures. We estimated the role of vasoconstriction in determining the level of pulmonary hypertension. This study included 26 patients with moderate pulmonary hypertension mean pulmonary arterial pressure (MPAP) = 27.3 +/- 1.2 mmHg) secondary to chronic obstructive lung disease (COLD), forced expiratory volume in one second (FEV1) = 0.95 +/- 0.13 l; arterial oxygen tension (PaO2) = 8.7 +/- 0.25 kPa). After insertion of a thermodilution catheter in the pulmonary artery and a cannula in the femoral artery, mixtures containing 15, 21, 30 and 100% oxygen were randomly administered for 20 min each. As fractional inspiratory oxygen (FIO2) increased, MPAP decreased relatively less than cardiac index. Cardiac output was at its highest during room air breathing and the hypoxic mixture did not lead to a further increase. Unlike normal subjects, in whom adjustment of cardiac output is achieved by heart rate alone, haemodynamic regulation in these patients also involved stroke volume. Variations in MPAP and cardiac index were strongly correlated with arterial oxygen saturation (SaO2). The greatest variations were noted in the patients with the highest pulmonary hypertension. Under normoxic and hyperoxic condition the relationship between pulmonary artery driving pressure and cardiac index was linear and its slope steeper in patients having the highest pulmonary hypertension at steady-state.(ABSTRACT TRUNCATED AT 250 WORDS)     

The use of oxygen and prostacyclin as pulmonary vasodilators in congenital heart disease

We have obtained dose-response curves for the effects of prostacyclin on the pulmonary and systemic circulations of 10 patients with pulmonary hypertension secondary to congenital heart disease. With the subjects breathing air, prostacyclin caused statistically significant, pulmonary and systemic vasodilation. When the patients breathed 100% oxygen, pulmonary blood flow rose and pulmonary vascular resistance fell with no significant change in pulmonary artery pressure. Prostacyclin had a small additional vasodilator effect, but this did not reach statistical significance. Prostacyclin may be of some use in the assessment of the reversibility of an elevated pulmonary vascular resistance when surgery for the underlying congenital heart defect is being contemplated.     

Does prostacyclin enhance the selective pulmonary vasodilator effect of oxygen in children with congenital heart disease?

We have obtained dose-response curves for the effects of prostacyclin on the pulmonary and systemic circulations in 20 children (median age 3 years) with pulmonary hypertension complicating congenital heart disease. Results were obtained with the children breathing both air and 100% oxygen. Under both sets of conditions, remote respiratory mass spectrometry was used to measure oxygen consumption and hence cardiac output by the direct Fick principle. When the subjects breathed air, prostacyclin caused a dose-dependent fall in pulmonary vascular resistance (measured in mm Hg . liter-1 . min . m2) (11.12 to 8.07, standard error of difference [SED] = 0.5, p less than .01). The level of the pulmonary vascular resistance when the subjects breathed air during the infusion of 20 ng/kg/min prostacyclin was not significantly different from that found when they breathed 100% oxygen and did not receive the drug (8.67 vs 8.93, SED = 0.55, p = NS). When infused while the subjects breathed 100% oxygen, prostacyclin caused additional dose-dependent pulmonary vasodilation (pulmonary vascular resistance 8.93 to 7.23, SED = 0.3, p less than .01). Unlike 100% oxygen, prostacyclin was not selective, and caused tachycardia and systemic hypotension at the higher doses. These results suggest that in children with congenital heart disease 100% oxygen does not maximally vasodilate the pulmonary circulation, and further pulmonary vasodilatation can be obtained with a blood-borne agent.     

Continuous low dose inhaled nitric oxide for treatment of severe pulmonary hypertension after cardiac surgery in paediatric patients.

OBJECTIVE--To assess the effect of inhaled nitric oxide (NO) on severe postoperative pulmonary hypertension in children after surgical repair of a congenital heart defect. DESIGN--A pilot study of NO administration to 7 consecutive children who required adrenergic support and in whom postoperative mean pulmonary artery pressure was more than two thirds of mean systemic pressure and persisted despite alkalotic hyperventilation. SETTING--Routine care after cardiac surgery for congenital heart disease in a multidisciplinary paediatric intensive care unit. METHODS--Continuous inhalation of NO, initially at 15 ppm. Therefore, daily attempts at complete weaning or at reducing NO to the lowest effective dose. RESULTS--In 6 of the 7 children NO inhalation selectively decreased mean (SD) pulmonary artery pressure from 51 (12) to 31 (9) mm Hg (P < 0.05) while mean systemic arterial pressure was unchanged (68 (10) v 71 (7) mm Hg) (NS) and the arteriovenous difference in oxygen content decreased from 6.7 (0.9) to 4.8 (0.8) vol% (P < 0.05). Concomitantly PaO2 increased from 158 (98) to 231 (79) mm Hg) (P < 0.05). The seventh child showed no response to NO up to 80 ppm, could not be weaned from cardiopulmonary bypass, and died in the operating room. In responders, attempts at early weaning from NO inhalation always failed and NO at concentrations of less than 10 ppm was continuously administered for a median of 9.5 days (range 4 to 16 days) until complete weaning was possible from a mean dose of 3.9 (2.9) ppm. Methaemoglobinaemia remained below 2% and nitrogen dioxide concentrations usually ranged from 0.1 to 0.2 ppm. One child later died and five were discharged. A few months after surgery Doppler echocardiography (and catheterisation in one) showed evidence of regression of pulmonary hypertension in all 5. CONCLUSIONS--Inhalation of NO reduced pulmonary artery pressure in children with severe pulmonary hypertension after cardiac surgery and this effect was maintained over several days at concentrations carrying little risk of toxicity.     

血管内超声评价先天性心脏病合并肺动脉高压肺血管病理变化的初步探讨

目的 :探索血管内超声在合并肺动脉高压的先天性心脏病 (先心病 )患儿肺动脉内检查的可行性并评价肺血管病理变化。　　方法 :对 6例合并肺动脉高压的先心病患儿进行肺动脉血管内超声检查 ,观测其管腔大小、内膜和中层的厚度以及血管的搏动性 ,同时与右心导管血液动力学在吸氧前后的变化结果进行了比较。　　结果 :肺动脉内超声能够清晰观测到其管壁的三层组织学变化 ,中层增厚 ,吸氧后肺动脉搏动性明显增强 ,与右心导管血液动力学检查结果有高度相关性。　　结论 :血管内超声对肺动脉的检查安全可靠 ,检查结果能够较好的反映肺动脉形态学变化。     

Factors influencing systemic arterial oxygen saturation in complete transposition of the great arteries

Hemodynamic data from 100 patients with complete transposition of the great arteries, all 1 year old or older, were reviewed. Only 2 patients, at rest and breathing room air, had a systemic arterial oxygen saturation level as high as 85 percent. In any patient with dextrotransposition of the great arteries and a systemic arterial saturation of 85 percent or greater, there is a strong likelihood of a single ventricle.

At a given level of oxygen consumption, 3 potential variables influence systemic arterial oxygen saturation in transposition: (1) magnitude of the intercirculatory mixing, (2) hemoglobin concentration and (3) magnitude of the recirculated systemic mixed venous flow. The first 2 factors are directly related to systemic saturation, and the third is inversely related.

Decreasing total pulmonary blood flow, secondary to increasing pulmonary vascular disease or pulmonary stenosis, results in decreased intercirculatory mixing and a decrease in systemic arterial oxygen saturation. This decrease in saturation is partially compensated for by an increase in hemoglobin concentration. An index calculated from the level of systemic arterial oxygen saturation divided by the value for hemoglobin concentration is useful in follow-up studies to determine the presence of pulmonary vascular disease or pulmonary stenosis. This index will often eliminate the need for repeated cardiac catheterizations before definitive corrective surgery is carried out.     

Arterial hypoxemia following the administration of sublingual nitroglycerin.

The effect of nitroglycerin on arterial blood gases and cardiovascular hemodynamics were studied in patients with coronary artery disease. In 13 premedicated patients blood gases and cardiovascular hemodynamics were studied before and 10 minutes after sublingual nitroglycerin (0.6 mg). In eight unpremedicated patients only blood gases were determined before and 10 minutes after sublingual nitroglycerin 0.6 mg. All studies were performed before induction of anesthesia with the patients in the supine position breathing room air. In both groups, arterial PO2 decreased significantly (p less than 0.001); pH and pCO2 did not change. In the 13 patients on which hemodynamic studies were performed, the mean arterial pressure (p less than 0.001), cardiac index (p less than 0.001), central venous pressure (p less than 0.001), pulmonary artery (p less than 0.001) and pulmonary artery wedge pressure (p less than 0.001) decreased. Calculated values for systemic and pulmonary vascular resistance were not significantly altered (p greater than 0.4). This study gives conclusive evidence that nitroglycerin reduces arterial PO2 in most patients with coronary artery disease breathing room air in the supine position. The possible mechanisms and clinical implications are discussed.     

Aerosolized iloprost for pulmonary vasoreactivity testing in children with long‐standing pulmonary hypertension related to congenital heart disease

Background: In congenital heart disease with increased pulmonary blood flow and pressure, progressive changes in the vascular structure can lead to irreversible pulmonary hypertension (PH). Pulmonary hemodynamic parameters are used to determine whether surgical correction is no longer indicated. In this study, aerosolized iloprost was used to assess pulmonary vasoreactivity in children with long‐standing PH related to congenital heart disease. Methods: Children with long‐standing and severe PH secondary to congenital heart disease were included in this study. Various hemodynamic parameters were measured before and after iloprost inhalation (0.5 μg/kg), and vascular resistance was determined. Responders to the iloprost test were defined as those with a decrease in both pulmonary vascular resistance (PVR) and pulmonary‐to‐systemic vascular resistance ratio (R_p/R_s) of >10%. Results: Eighteen children aged between 7 months and 13 years with long‐standing and severe PH secondary to congenital heart disease were studied. Thirteen children had a positive response, resulting in a mean (± SD) decrease of PVR from 9.3 ± 4.6 to 4.6 ± 2.7 Wood U · m² (P < 0.001), and a mean decrease of R_p/R_s from 0.54 ± 0.37 to 0.24 ± 0.14 (P = 0.005). Conclusions: Iloprost‐induced pulmonary vasodilator responses vary among children with PH related to congenital heart disease. The use of inhale iloprost in the cardiac catheterization laboratory results in pulmonary vasoreactivity for some of these children particularly a reduction in PVR and the pulmonary‐to‐systemic vascular resistance ratio. © 2008 Wiley‐Liss, Inc.     

Effect of inhaled nitric oxide on raised pulmonary vascular resistance in children with congenital heart disease.

OBJECTIVE--To study the short-term effects of inhaled nitric oxide in infants and young children with congenital heart disease. SETTING--A supraregional referral centre for children with congenital heart disease. PATIENTS AND METHODS--22 infants and children aged 3-32 months (median age 5 months) with congenital heart disease undergoing preoperative cardiac catheterisation. All but one infant had intracardiac shunt lesions and 13 had increased pulmonary vascular resistance. During catheterisation the patients inhaled nitric oxide in a concentration of 40 parts per million in room air. Pulmonary and systemic haemodynamic variables were evaluated by means of measured oxygen consumption and the Fick principle before and after 10 minutes' exposure to nitric oxide. RESULTS--Inhaled nitric oxide did not affect the systemic circulation. There was a significant reduction in the pulmonary vascular resistance, but only in the 13 infants with pulmonary hypertension, in whom pulmonary vascular resistance was reduced by 34% from 8.6 (4.6) mm Hg.min.m2.l-1 (mean (SD)) to 5.7 (3.5) mm Hg.min.m2.l-1. The pulmonary circulation in infants with normal pulmonary vascular resistance was not affected. No statistically significant increase in methaemoglobin was seen, though there were large individual differences. No other side effects were seen. CONCLUSION--The present study shows that in infants with congenital heart disease inhaled nitric oxide reduced pathologically increased pulmonary vascular resistance without affecting systemic circulation and without important side effects with brief exposure.     

Effect of long term oxygen treatment at home in children with pulmonary vascular disease.

Inhalation of 100% oxygen by nine children with pulmonary vascular disease increased pulmonary blood flow measured at cardiac catheterisation; there was no significant change in pulmonary artery pressure. Fifteen children with pulmonary vascular disease that was severe enough to preclude corrective cardiac operation were studied to determine the effect of long term oxygen treatment on pulmonary vascular disease. Nine received long term domiciliary oxygen for a minimum of twelve hours a day for up to five years. Though the untreated group closely resembled the treated group their survival was significantly less good. All nine treated children are alive whereas five of the six children who did not receive oxygen have died.     

Responsiveness of raised pulmonary vascular resistance to oxygen assessed by pulsed Doppler echocardiography.

OBJECTIVE--To assess whether changes in Doppler echocardiographic indices in the pulmonary artery correlated with changes in pulmonary vascular resistance. DESIGN--Acceleration time, ejection time, maximal flow velocity, and velocity time integrals were measured at the same time as pressure and oxygen saturation measurements in room air and during 10 minutes of oxygen breathing in the catheterisation laboratory. Pulmonary vascular resistance and pulmonary blood flow (Qp) were calculated from catheterisation data by use of the Fick principle. PATIENTS--14 consecutive patients with a congenital heart defect and a left to right shunt associated with raised pulmonary artery pressure who underwent routine diagnostic cardiac catheterisation to assess their pulmonary vascular resistance. RESULTS--Though pulmonary vascular resistance and systolic pulmonary artery pressure fell significantly during oxygen administration, there was no significant change in the acceleration time or ejection time. Peak velocity increased significantly during oxygen administration. During oxygen breathing Doppler derived measurements of pulmonary flow showed a significant increase in Qp similar to the increase in Qp measured by the Fick principle. There was no significant correlation between the fall in pulmonary vascular resistance and the increase in acceleration time or ejection time, increase in peak velocity, increase in pulmonary artery diameter, or increase in Doppler derived pulmonary blood flow. CONCLUSIONS--Measurements of acceleration and ejection time by Doppler echocardiography did not predict the response of pulmonary artery pressure and resistance to oxygen. Though changes in maximal flow velocity across the pulmonary artery and in Doppler derived pulmonary blood flow measurements became significant during oxygen breathing, the correlation of these changes with fall in pulmonary vascular resistance was poor.     

Responsiveness of raised pulmonary vascular resistance to oxygen assessed by pulsed Doppler echocardiography.

OBJECTIVE--To assess whether changes in Doppler echocardiographic indices in the pulmonary artery correlated with changes in pulmonary vascular resistance. DESIGN--Acceleration time, ejection time, maximal flow velocity, and velocity time integrals were measured at the same time as pressure and oxygen saturation measurements in room air and during 10 minutes of oxygen breathing in the catheterisation laboratory. Pulmonary vascular resistance and pulmonary blood flow (Qp) were calculated from catheterisation data by use of the Fick principle. PATIENTS--14 consecutive patients with a congenital heart defect and a left to right shunt associated with raised pulmonary artery pressure who underwent routine diagnostic cardiac catheterisation to assess their pulmonary vascular resistance. RESULTS--Though pulmonary vascular resistance and systolic pulmonary artery pressure fell significantly during oxygen administration, there was no significant change in the acceleration time or ejection time. Peak velocity increased significantly during oxygen administration. During oxygen breathing Doppler derived measurements of pulmonary flow showed a significant increase in Qp similar to the increase in Qp measured by the Fick principle. There was no significant correlation between the fall in pulmonary vascular resistance and the increase in acceleration time or ejection time, increase in peak velocity, increase in pulmonary artery diameter, or increase in Doppler derived pulmonary blood flow. CONCLUSIONS--Measurements of acceleration and ejection time by Doppler echocardiography did not predict the response of pulmonary artery pressure and resistance to oxygen. Though changes in maximal flow velocity across the pulmonary artery and in Doppler derived pulmonary blood flow measurements became significant during oxygen breathing, the correlation of these changes with fall in pulmonary vascular resistance was poor.     

Pulmonary vascular extraction of circulating norepinephrine in infants with bronchopulmonary dysplasia

Although the pulmonary circulation in infants with advanced bronchopulmonary dysplasia (BPD) is characterized by abnormal structure and vasoreactivity, metabolic lung functions have not been studied in these infants. To test the hypothesis that patients with severe BPD may have abnormal metabolic lung function, we assessed the pulmonary vascular extraction of circulating norepinephrine in six children with BPD during cardiac catheterization. Plasma norepinephrine levels were measured from simultaneously drawn mixed venous (main pulmonary artery) and left atrium or femoral artery samples. In comparison with four infants with mild heart disease without pulmonary hypertension, we found that infants with BPD extract proportionately less norepinephrine than non-BPD infants [-7 +/- 50% (BPD) versus +27 +/- 6% (non-BPD); P less than 0.001, t test]. Three infants with BPD had higher arterial than mixed venous concentrations of plasma norepinephrine, suggesting net production across the lung. Plasma catecholamine levels and percent extraction correlated poorly with cardiac index and systemic and pulmonary vascular resistance indices. However, this study group was characterized by a high incidence of pulmonary (6/6) and systemic (4/6) hypertension, left ventricular hypertrophy (4/6), and subsequent death (3/6). We conclude that infants with severe BPD and pulmonary hypertension have decreased pulmonary vascular clearance or net production of circulating norepinephrine, but links between altered pulmonary catecholamine metabolism and pulmonary hypertension, or other cardiovascular abnormalities associated with BPD, remain speculative.     

Acute effects of oxygen, nifedipine, and diltiazem in patients with cystic fibrosis and mild pulmonary hypertension

The acute effects of oxygen, nifedipine, and diltiazem were studied in eight patients with cystic fibrosis and mild pulmonary hypertension, to assess the possibility of relieving the latter before the occurrence of irreversible vascular changes. Oxygen decreased pulmonary pressure (-23%) and resistance (-21%), while increasing systemic resistance (+23%). Nifedipine increased cardiac index (+30%), at the expense of augmented right ventricular work (+42%), resulting in a decreased calculated pulmonary resistance (-23%); pulmonary artery pressure remained unchanged, however. Nifedipine decreased arterial Po2 (-10%), suggesting ventilation-perfusion mismatch. Four of the eight patients responded to diltiazem. Their pulmonary pressure (-35%) and resistance (-43%) decreased, while systemic vascular tone remained unchanged. Oxygen in three patients, and diltiazem in two, returned pulmonary pressures and resistances to normal values. Early reversal of pulmonary hypertension is possible, and intervention is desirable before the establishment of chronic hypoxia, cor pulmonale, or right ventricular failure. Our data does not support the use of nifedipine in pulmonary hypertension, but shows that oxygen, and in some cases diltiazem, act as effective and selective pulmonary vasodilators.     

Hemodynamic effects of dobutamine in children.

Dobutamine is useful for augmenting cardiovasuclar function in adults. However, no information is available on the action of dobutamine in children. To determine its hemodynamic effects in children, we infused dobutamine into 12 children with congenital heart disease during diagnostic cardiac catheterization. We administered dobutamine in two doses: first 2 and then 7.75 microgram/kg per min for 10 minutes each. We meaured heart rate, cardiac output, systemic and pulmonary arterial, right atrial and pulmonary capillary blood pressures before and during the infusion of dobutamine. Systemic and pulmonary vascular resistances, cardiac index and stroke index were calculated. Cardiac output, cardiac index, stroke volume, stroke index and systemic arterial phasic and mean blood pressures increased sugnificantly (P less than 0.05) and pulmonary capillary mean blood pressure decreased significantly (P less than 0.05) during the infusion of each dose of dobutamine compared with control values. Heart rate, pulmonary and right atrial mean blood pressure and systemic and pulmonary vascular resistance were unchanged with either dose of dobutamine. We noted no adverse effect from the drug.     

Effect of long term oxygen treatment at home in children with pulmonary vascular disease

Inhalation of 100% oxygen by nine children with pulmonary vascular disease increased pulmonary blood flow measured at cardiac catheterisation; there was no significant change in pulmonary artery pressure. Fifteen children with pulmonary vascular disease that was severe enough to preclude corrective cardiac operation were studied to determine the effect of long term oxygen treatment on pulmonary vascular disease. Nine received long term domiciliary oxygen for a minimum of twelve hours a day for up to five years. Though the untreated group closely resembled the treated group their survival was significantly less good. All nine treated children are alive whereas five of the six children who did not receive oxygen have died.     

Effects of removing oxygen from patients with chronic obstructive pulmonary disease

To determine the acute physiologic effects of removing oxygen from patients with chronic obstructive pulmonary disease (COPD) who are receiving long-term oxygen therapy, we made serial measurements in 20 patients during and after stopping low-flow oxygen therapy. Removing oxygen caused an increase in pulmonary vascular resistance, requiring 2 to 3 h to reach a new steady state. Removing oxygen therapy increased pulmonary vascular resistance index (PVRI) by 31% during rest (8.14 +/- 0.61 versus 6.23 +/- 0.51 units, p less than 0.001) and by 29% during exercise (8.11 +/- 0.9 versus 6.31 +/- 0.7, p less than 0.001). The increase in PVRI occurred because of an increase in pulmonary arterial pressure without a change in pulmonary capillary wedge pressure or cardiac index. At rest the increase in pulmonary arterial pressure caused by stopping oxygen correlated with the decrease in arterial oxygen saturation (r = 0.70, p less than 0.01). Removing oxygen decreased stroke volume index during rest and exercise. Although removing oxygen increased pulmonary vascular resistance, it did not affect systemic arterial pressure or vascular resistance. Stopping oxygen reduced arterial and mixed venous oxygen tension and oxygen delivery during rest and exercise. In patients who had a normal PaCO2 while breathing room air, removing oxygen therapy increased their oxygen consumption; conversely, in those patients who had an elevated PaCO2 while breathing room air, stopping oxygen therapy reduced oxygen delivery and oxygen consumption.(ABSTRACT TRUNCATED AT 250 WORDS)