• cerebral oxygenation;
  • chronic obstructive pulmonary disease;
  • exercise;
  • lung physiology;
  • near-infrared spectroscopy;
  • non-invasive ventilation



It is currently unknown whether potential haemodynamic improvements induced by non-invasive ventilation (NIV) would positively impact upon cerebral oxygenation (COx) in patients with moderate-to-severe chronic obstructive pulmonary disease (COPD).


To investigate the effects of NIV on exercise COx in COPD patients presenting with exercise-related O2 desaturation.


On a double-blind trial, 13 males (FEV= 48·8 ± 15·1% predicted) were randomly assigned to NIV (16 cmH2O IPS and 5 cmH2O PEEP) plus HOx (FiO=  0·4) or sham NIV (7 cmH2O IPS and 5 cmH2O PEEP to overcome breathing circuit resistance) plus HOx during ramp-incremental exercise performed on different days. Near-infrared spectroscopy and impedance cardiography assessed changes (Δ) in COx and cardiac output (QT), respectively.


There were no significant between-intervention differences in peak work rate, ventilation and reported symptoms (P>0·05). Peripheral oxyhaemoglobin saturation remained above 98% throughout the tests. NIV + HOx was associated with larger increases in Δ COx, Δ QT and Δ stroke volume at maximal and submaximal exercise (P<0·05). Increases in the area under the curve (to an iso-work rate) of Δ COx under NIV + HOx were significantly (P<0·01) correlated with improvements in Δ QT (r = 0·82) and Δ stroke volume (r = 0·87). There was, however, no significant correlation between enhancement in these physiological responses with changes in peak work rate with NIV + HOx (P>0·05).


NIV added benefit to HOx in improving central haemodynamics and COx in O2 ‘desaturators’ with COPD. The clinical relevance of such beneficial effects on exercise tolerance, however, remains to be demonstrated.