Background The effects of anesthetics on the injured brain continue to be the subject of controversial discussion. Since isoflurane has recently been shown to induce apoptosis of cerebral endothelial cells, this study compared different anesthetic compounds regarding their potential to induce cerebro-vascular apoptosis. Methods The in vitro model of the blood-brain barrier used in this study consisted of astrocyte-conditioned human umbilical vein endothelial cells (AC-HUVEC) has been used. After 24 h of deep hypoxia and reoxygenation or control treatment, AC-HUVEC were exposed to 0, 0.5, 1.0, or 2.0 times the minimum alveolar concentration of isoflurane or sevoflurane, or 0, 75, 150, or 300 nM of midazolam for 2 h. After 24 h, AC-HUVEC were harvested, and the degree of apoptosis was assessed by means of Western blots for the Bax and Bcl-2 ratio and, for controls and the highest concentration groups, terminal deoxynucleotidyl-mediated dUTP-biotin nick end labeling (TUNEL). Results Without hypoxic pretreatment, 2.0 MAC of isoflurane slightly increased TUNEL intensity compared to control and sevoflurane, but without any significant changes in the Bax and Bcl-2 ratio. After hypoxic pretreatment, exposure to isoflurane led to a multifold increase in the Bax and Bcl-2 ratio in a dose dependent manner, which was also significantly higher than the ratio observed in the 2 MAC sevoflurane group. TUNEL intensity in the post-hypoxic 2 MAC isoflurane group was increased by a factor of 11 vs. control and by 40 vs. sevoflurane. Sevoflurane and midazolam did not significantly alter these markers of apoptosis, when compared to the control group. Conclusions Isoflurane administered after hypoxia elevates markers of apoptosis in endothelial cells transdifferentiated to the cerebro-vascular endothelium. Endothelial apoptosis may be a previously underestimated mechanism of anesthetic neurotoxicity. Administration of high concentrations of isoflurane in experimental settings may have negative effects on the blood-brain barrier.