To the Editor: We read with interest the article by Kaps et al1 concerning the nature of Doppler-detected microembolic signals (MES) in the cerebral arteries of patients with an artificial prosthetic heart valve. Their oxygen experiment provides strong evidence for the conclusion that these MES are gas bubbles. They noted a strong reduction in the number of MES when patients breathed 100% oxygen instead of air (21% oxygen) at atmospheric pressure (ie, 100 kPa and not 1 kPa, as mentioned in the article). The nitrogen present in air bubbles is replaced by oxygen, and because oxygen bubbles have a shorter life span because of higher solubility in blood, the number of MES reaching the brain is reduced. Up to this point, we agree with the authors. Kaps et al also performed a hyperbaric chamber experiment (n=1). Compared with the baseline condition (air respiration at atmospheric pressure), they noted a strong reduction in the number of MES when the patient inhaled 100% oxygen at an increased ambient pressure of 175 kPa, an intermediate reduction in the number of MES under …