Right-to-left shunts (RLS) are prevalent in patients with chronic obstructive pulmonary disease (COPD) and might exaggerate oxygen desaturation, especially at altitude. The aim of this study was to describe the prevalence of RLS in patients with COPD traveling to altitude and the effect of preventive dexamethasone. Lowlanders with COPD [Global Initiative for Chronic Obstructive Lung Disease (GOLD) grades 1–2, oxygen saturation assessed by pulse oximetry ([Formula: see text]) >92%] were randomized to dexamethasone (4 mg bid) or placebo starting 24 h before ascent from 760 m and while staying at 3,100 m for 48 h. Saline-contrast echocardiography was performed at 760 m and after the first night at altitude. Of 87 patients (81 men, 6 women; mean ± SD age 57 ± 9 yr, forced expiratory volume in 1 s 89 ± 22% pred, [Formula: see text] 95 ± 2%), 39 were assigned to placebo and 48 to dexamethasone. In the placebo group, 19 patients (49%) had RLS, of which 13 were intracardiac. In the dexamethasone group 23 patients (48%) had RLS, of which 11 were intracardiac ( P = 1.0 vs. dexamethasone). Eleven patients receiving placebo and 13 receiving dexamethasone developed new RLS at altitude ( P = 0.011 for both changes, P = 0.411 between groups). RLS prevalence at 3,100 m was 30 (77%) in the placebo and 36 (75%) in the dexamethasone group ( P = not significant). Development of RLS at altitude could be predicted at lowland by a higher resting pulmonary artery pressure, a lower arterial partial pressure of oxygen, and a greater oxygen desaturation during exercise but not by treatment allocation. Almost half of lowlanders with COPD revealed RLS near sea level, and this proportion significantly increased to about three-fourths when traveling to 3,100 m irrespective of dexamethasone prophylaxis. NEW & NOTEWORTHY The prevalence of intracardiac and intrapulmonary right-to-left shunts (RLS) at altitude in patients with chronic obstructive pulmonary disease (COPD) has not been studied so far. In a large cohort of patients with moderate COPD, our randomized trial showed that the prevalence of RLS increased from 48% at 760 m to 75% at 3,100 m in patients taking placebo. Preventive treatment with dexamethasone did not significantly reduce the altitude-induced recruitment of RLS. Development of RLS at 3,100 m could be predicted at 760 m by a higher resting pulmonary artery pressure and arterial partial pressure of oxygen and a more pronounced oxygen desaturation during exercise. Dexamethasone did not modify the RLS prevalence at 3,100 m.