Abstract
The aim of our study was to find out how blood gas disturbances in stable, eucapnic, severe chronic obstructive pulmonary disease (COPD) patients with an arterial oxygen tension (PaO2) value of 7·7 (6·1–8·4) kPa are affected by ventilation–perfusion (VA/Q) relationships and carbon dioxide (CO2) sensitivity and how these parameters are influenced by 6 months of long‐term oxygen treatment (LTOT). VA/Q ratios were measured using the multiple inert gas elimination technique (MIGET). Mouth occlusion pressure 0·1 s after onset of inspiration (Pi0·1) and minute ventilation (VE) were measured to assess respiratory drive response (ΔPi0·1/ΔPCO2) and hypercapnic ventilatory response (HCVR) to CO2 rebreathing. At the start of LTOT, a normal median respiratory drive response level of 1·2 (0·2–2·3) cm H2O/kPa and a low median HCVR as compared with healthy individuals (PE, was directed towards hypoperfused lung areas. The dispersion of ventilation (log SDV; 0·47–1·76), and the dispersion of perfusion (log SDQ; 0·66–1·07) were wider than normal. The PaO2 level correlated inversely with mean VA/Q ratio for ventilation (V mean) and shunt. The PaCO2 level correlated inversely with HCVR and vital capacity. After 6 months of LTOT, no significant changes in daytime blood gas levels, CO2‐sensitivity or VA/Q ratios were found. VE tended to be reduced by 1·0 l min–1. Conclusions: An elevated V mean and probably shunting are important contributing factors for the reduced PaO2 and hypercapnic ventilatory response is a major determinant of PaCO2 in eucapnic stable hypoxaemic COPD. Six months of LTOT does not affect blood gases, CO2 sensitivity or ventilation–perfusion relationships.

This publication has 28 references indexed in Scilit: