Moderate concentrations of supplemental oxygen worsen hypercapnia in obesity hypoventilation syndrome: a randomised crossover study.

Abstract

In people with obesity hypoventilation syndrome (OHS), breathing 100% oxygen increases carbon dioxide (PCO2), but its effect on pH is unknown. This study investigated the effects of moderate concentrations of supplemental oxygen on PCO2, pH, minute ventilation (VE) and physiological dead space to tidal volume ratio (VD/VT) among people with stable untreated OHS, with comparison to healthy controls. In a double-blind randomised crossover study, participants breathed oxygen concentrations (FiO2) 0.28 and 0.50, each for 20 min, separated by a 45 min washout period. Arterialised-venous PCO2 (PavCO2) and pH, VE and VD/VT were measured at baseline, then every 5 min. Data were analysed using general linear model analysis. 28 participants were recruited (14 OHS, 14 controls). Among OHS participants (mean ± SD arterial PCO2 6.7 ± 0.5 kPa; arterial oxygen 8.9 ± 1.4 kPa) FiO2 0.28 and 0.50 maintained oxygen saturation 98-100%. After 20 min of FiO2 0.28, PavCO2 change (ΔPavCO2) was 0.3 ± 0.2 kPa (p = 0.013), with minimal change in VE and rises in VD/VT of 1 ± 5% (p = 0.012). FiO2 0.50 increased PavCO2 by 0.5 ± 0.4 kPa (p = 0.012), induced acidaemia and increased VD/VT by 3 ± 3% (p = 0.012). VE fell by 1.2 ± 2.1 L/min within 5 min then recovered individually to varying degrees. A negative correlation between ΔVE and ΔPavCO2 (r = -0.60, p = 0.024) suggested that ventilatory responses were the key determinant of PavCO2 rises. Among controls, FiO2 0.28 and 0.50 did not change PavCO2 or pH, but FiO2 0.50 significantly increased VE and VD/VT. Commonly used oxygen concentrations caused hypoventilation, PavCO2 rises and acidaemia among people with stable OHS. This highlights the potential dangers of this common intervention in this group.