This study defines the vascular territories of the nasal fossa arteries and includes photographs of dissections.

Epistaxis is a common and in most cases benign event. Severe or recurrent epistaxis, however, can present therapeutic problems. Forty-five cases of supraselective embolization in intractable epistaxis are reported. The authors' success rate of 97%, similar to the success rates reported in the literature, confirms the effectiveness of this technique. Complications occurred in only 8% of the cases. One serious complication was neurologic. Percutaneous embolization is an effective option for managing intractable posterior nasal bleeding but is not recommended as an early form of treatment.

The stapedial artery is an embryonic artery which disappears during the tenth week in utero, in human species. During its short life, this artery shapes the stapes and transforms the middle meningeal artery from the internal carotid artery to a branch of the external carotid system. Nevertheless, a persistent stapedial artery is seen in 0.2-4.8 per thousand of human adults. This persistence is usually asymptomatic but can sometimes cause pulsatile tinnitus or conductive hearing loss. Despite the risk of facial palsy, hearing loss and even hemiplegia argued by several authors, some surgeons have succeeded in coagulation without side effects. Reviewing the literature, we seek to enlighten the actual knowledge about the persistent stapedial artery to evaluate the risk to coagulate it. Embryologic studies explain the four types of persistent stapedial arteries: the hyoido-stapedial artery, the pharyngo-stapedial artery, the pharyngo-hyo-stapedial artery and aberrant internal carotid with persistent stapedial artery. Phylogenetic studies show that the stapedial artery persists in adulthood in many vertebrates. Its disappearance is therefore either a random effect or an adaptative convergence. This adaptation could be partially linked to the negative allometry of the stapes. Practically, the risk to coagulate a stapedial artery seems limited thanks to anastomoses, for example with the stylomastoid artery. The risk of hemiplegia reported is in fact an extrapolation of variation in rats' embryos. A persistent stapedial artery can therefore reasonably be coagulated, with special attention to the facial nerve, because the facial canal is always dehiscent where the artery penetrates.