Juvenile nasopharyngeal angiofibroma often attaches firmly to the adjoining bony region around the sphenopalatine foramina–sphenopalatine fossa–pterygomaxillary fissure. This can result in hourglass-shaped constriction and predispose to incomplete resection (residual disease) with a transpalatal approach. This paper describes attempts to address this ‘inaccessible’ area with a novel instrument, used since 2012.

Measurements of the sphenopalatine foramen, nasal septum, posterior nasopharyngeal wall and hard palate were undertaken in 20 skulls and 10 computed tomography scans (lateral extension). A device was designed (in terms of angulation and length) following several trials with malleable wire. A search of patents was also undertaken. Recurrence rates were compared in cases of device use and non-use.

The novelty of the sphenopalatine fossa dissector was established and the device was patented. This device has significantly improved our 17.59 per cent recurrence rate of the past 4 decades; of 63 cases over 3 years, there were only 3 recurrences and 2 residual disease cases. Findings of our previous studies with or without the device are compared.

Existing evidence supports the incorporation of this inexpensive instrument in the armamentarium for resecting lateral extension of juvenile nasopharyngeal angiofibroma during a transpalatal approach.

To establish whether nasal bony landmarks on computed tomography could be utilised reliably in endoscopic approaches to the sphenopalatine foramen.

A prospective analysis of 102 consecutive helical computed tomography scans of the paranasal sinuses was carried out by 2 senior ENT surgeons. Distances from the sphenopalatine foramen to endoscopically palpable bony landmarks were measured.

There were a total of 102 patients (45 females and 57 males), with a mean age of 62 years. The mean distance from the posterior fontanelle to the sphenopalatine foramen was 14.1 mm (standard deviation = 2.13). The average vertical distance of the sphenopalatine foramen opening from the bony attachment of the inferior turbinate was 14.13 mm. There were no statistically significant differences between any of these measurements (foramen width p-value = 0.714, distance from fontanelle p-value = 0.43 and distance from inferior turbinate p-value = 0.48).

Determination of reliable bony landmarks is clinically useful in endoscopic surgery and can aid identification of the sphenopalatine foramen. The inferior turbinate concha and posterior fontanelle may be used as reliable computed tomography landmarks for endoscopic approaches to the sphenopalatine foramen.

To identify measurements that may help intra-operative localisation of the sphenopalatine foramen.

The study used three-dimensional surgical navigation software to study radiological anatomy, in order to define the distances and angulations between identifiable bony landmarks and the sphenopalatine foramen.

The distance from the anterior nasal spine to the sphenopalatine foramen was 59 mm (±4 mm; inter-observer variation = 0.866; intra-observer variation = 0.822). The distance from the piriform aperture to the sphenopalatine foramen was 48 mm (±4 mm; inter-observer variation = 0.828; intra-observer variation = 0.779). The angle of elevation from the nasal floor to the sphenopalatine foramen was 22° (±3°; inter-observer variation = 0.441; intra-observer variation = 0.499).

The sphenopalatine foramen is consistently identifiable on three-dimensional, reconstructed computed tomography scans. Repeatable measurements were obtained. The centre point of the foramen lies 59 mm from the anterior nasal spine at 22° elevation above the plane of the hard palate and 48 mm from the piriform aperture. We discuss how these data could be used to facilitate intra-operative location of the sphenopalatine foramen in difficult cases.