Correction of hypoxia and hypercapnia in COPD patients: effects on cerebrovascular flow.

To assess the responsiveness of cerebral blood flow to arterial carbon dioxide tension (Pa,CO2), arterial oxygen tension (Pa,O2), and pH modifications, in chronic hypercapnia, we measured middle cerebral blood flow velocity (CBFV) by transcranial Doppler ultrasound in 13 chronically hypercapnic, long-term ventilated patients with chronic obstructive pulmonary disease (COPD), in the following conditions: 1) breathing room air; 2) with oxygen supplementation; 3) during mechanical noninvasive intermittent positive pressure ventilation (nIPPV) with O2 enrichment. Under baseline conditions (room air), the CBFV was within the normal range. During oxygen administration, a statistically significant increase was obtained in Pa,O2 (6.5 +/- 0.6 vs 11.2 +/- 1.9 kPa (49.1 +/- 4.3 vs 84.3 +/- 14.6 mmHg)), without relevant variations in: CBFV (54.2 +/- 9.1 cm.s-1), Pa,CO2 (8.6 +/- 1.0 kPa (64.7 +/- 7.7 mmHg)) and hydrogen ion concentration [H+] (42.9 +/- 2.9 nM), compared to baseline values (CBFV = 52.8 +/- 10.7 cm.s-1; Pa,CO2 = (8.4 +/- 0.9 kPa (63.1 +/- 7.1 mmHg; [H+] = 41.8 +/- 2.8 nM). After nIPPV, Pa,O2 did not increase any further (10.6 +/- 1.7 kPa (79.2 +/- 12.7 mmHg)), while CBFV (40.9 +/- 12.6 cm.s-1), Pa,CO2 (7.5 +/- 1.3 kPa (56.2 +/- 9.4 mmHg)) and [H+] (39.1 +/- 4.6 nM) showed a significant reduction compared to oxygen therapy (p < 0.01). We therefore conclude that in chronically hypercapnic long-term ventilated patients cerebral blood flow depends mainly on changes in Pa,CO2 and [H+], whilst oxygen does not seem to interfere with cerebral flow velocity. The reduction of Pa,CO2, due to mechanical ventilation, may determine cerebral blood vessel constriction, with possible impairment of cerebral functions.

Clinical usefulness of a new portable oxygen concentrator.

Thirty chronic, hypoxaemic patients with a mean arterial oxygen tension (PaO2) of 6.81 kPa (SD 0.56) in air were tested using the Travelair. The study was performed at rest, allowing the patients to breathe in the following conditions: a) compressed air; b) continuous oxygen flow from the concentrator; c) oxygen from the concentrator in demand-valve mode (DV) with an activation time (AT) of 375 ms; d) DV with AT of 750 ms; e) DV with AT of 1,125 ms; f) DV with AT 1,500 ms; and g) continuous oxygen from the hospital outlet at 2 l.min-1. The mean (SD) SaO2% values at each consecutive step were: a) 87.2 (5.0)%; b) 93.0 (3.0)%; c) 93.9 (2.8)%; d) 94.2 (2.5)%; e) 94.0 (2.7)%; f) 94.1 (2.7)%; and g) 94.8 (2.4)%, respectively. Each of the results obtained with DV (b-f) was statistically different from those obtained breathing air (a) or oxygen (g). The mean (SD) respiratory rates at each consecutive step were: a) 21.2 (4.1); b) 21.0 (4.0); c) 21.3 (4.0); d) 21.0 (3.6); e) 20.7 (3.4); f) 21.0 (3.8); and g) 21.2 (3.8) breaths.min-1, respectively. No relationship was found between the mean oxygen saturations and the mean respiratory frequencies of the patients in each condition tested. With the concentrator on demand, the average of the best SaO2% obtained by the patients in whatever of the four DV activation time modes (conditions c-f) was 94.8 (2.4)%, and no statistical difference was detected between this result and the SaO2% obtained with 2 l.min-1 of continuous oxygen.