Transmaxillary Endoscopic Approach to the Lateral Recess of the Sphenoid Sinus: An Anatomical Study

Background Endonasal approaches to the lateral recess of the sphenoid sinus (LRSS) often include sacrifice of the ipsilateral vidian nerve (VN) and other pterygopalatine fossa (PPF) contents. This makes an alternative approach to the LRSS highly desirable, especially when the disease to be treated is benign, located far laterally (e.g., LRSS spontaneous cerebrospinal fluid leaks), and/or VN function preservation is necessary (e.g., patients with preoperative dry-eye symptoms; Fig. 1).

Objective To study the feasibility and establish the main anatomical landmarks of a purely transmaxillary endoscopic approach to the LRSS.

Methods A cadaveric dissection study was conducted. Only well-pneumatized LRSS were included, comprising six specimens and six sides in total. The exposure was initiated with an ipsilateral Caldwell-Luc approach (Fig. 2) and conducted using a 0-degree rod lens endoscope. The posterior wall of the maxillary sinus was widely removed and the PPF was carefully opened. Main anatomical landmarks were identified: the infraorbital nerve (ION) and the internal maxillary artery (IMAX). Blunt dissection between these structures was performed until foramen rotundum (FR) and the connections between the pterygopalatine ganglion (PPG) and V2 could be identified (Fig. 3). The periosteum was elevated to completely expose the base of the pterygoid process (BPP) and the greater wing of the sphenoid bone (GWSB). Bone drilling was performed until the lateral recess of the sphenoid sinus was widely opened.

Results The transmaxillary endoscopic approach to the LRSS was completed in all specimens. The ION was the most important anatomical landmark—it was easily identified and directly led to V2 and FR. For very well-pneumatized LRSS, ION had to be displaced superiorly to adequately expose the GWSB; this required removal of the bone overlying the infraorbital canal and the orbital process of the palatine bone. IMAX had to be removed in some specimens to increase the surgical exposure. With careful dissection of the PPF, it was possible to completely preserve V2, the PPG, and the connections between both structures. Drilling of the BPP and the GWSB was initiated just below FR and progressed laterally. Good surgical window was achieved in all specimens (superior limit: ION; medial limit: PPG and its connections to V2; inferior limit: IMAX, if preserved; lateral limit: infratemporal fossa muscles; Fig. 4). The most lateral part of the LRSS could be easily reached in all specimens (Fig. 5).

Conclusion The proposed transmaxillary endoscopic approach to the LRSS is feasible and permits direct access to the most lateral aspect of the LRSS, while preserving critical neural structures. Most important landmarks are the ION and the IMAX. The latter can be removed to increase surgical exposure.