A novel nasoseptal flap harvesting technique in revision expanded endoscopic transsphenoidal approaches
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.198214
Source of Support: None, Conflict of Interest: None
Objectives: To describe the technique of harvesting the nasoseptal flap (NSF) in revision-expanded endoscopic approaches (EEA).
Keywords: Cerebrospinal fluid leak, nasal septum, pituitary adenoma, sphenoid sinus, sphenopalatine artery
The introduction of the pedicled nasoseptal flap (NSF) was the most significant step that reduced the incidence of postoperative cerebrospinal fluid (CSF) leak from 20% to <5% following an expanded endoscopic approach (EEA) to the anterior skull base.,, However, in patients previously operated transnasally, defects in the septum and extensive fibrosis with adhesions might preclude the use of such flaps in revision cases. We present our experience in raising a flap based on the intact left sphenopalatine artery for successful endoscopic skull base reconstruction (ESBR), despite the prior posterior septectomy and transsphenoidal surgery.
Four patients with recurrent pituitary adenomas underwent revision EEA. The preoperative rigid nasal endoscopy, magnetic resonance imaging (MRI), and computed tomography (CT) scans were used to assess the status of the sphenopalatine pedicle, which was found to be intact on the left side [Figure 1]. All patients underwent an endoscopic transsphenoidal surgery via a binostril approach, and the CSF leak repair during the primary surgery was done using a fat graft. The NSF was not used in the primary surgery. During the revision surgery, the NSF was raised on the left side at the beginning of the surgery in all patients.
Following tumor excision, the dural defect was repaired with fat and fascia lata as inlay and onlay placements, respectively. The NSF was placed over the onlay fascia lata onto the bony edges of the defect. Surgicel, tissue glue, and gelfoam were placed sequentially over the repair site and bilateral merocel packs were kept for 5 days. The patient was kept on bedrest with a lumbar subarachnoid drain for 5 days. We endoscopically cleaned the nose twice in the first two weeks, following which the patients were asked to report for follow-up at 3 months.
Nasoseptal flap harvesting technique
The inferior incision is placed along the roof of the left choanae, comes on to the posterior half of the septum, drops down onto the lower edge of the nasal septum and comes forward along the floor of the nose, upto the free border of the septal cartilage [Figure 2].
The superior incision is the critical step wherein the strategic placement of the incision is done to preserve the left sphenopalatine pedicle. This incision can be considered in two parts – posterior half and the anterior half.
The posterior half starts on the left side along the sphenoid sinus floor edge, as high as possible, and connects with the previous sphenoidotomy opening at its upper margin [Figure 2]. Thereafter, the incision crosses the midline onto the right side of the previous sphenoidotomy opening and comes down, transecting the right sphenopalatine pedicle onto the roof of the right choana [Figure 2]. The incision then comes forward onto the posterior part of the right nasal septum and goes inferiorly to reach the floor of the nose on the right side [Figure 2]. The incision is then extended anteriorly along the floor of the right nose and then ascends upwards to reach the anterior edge of the septal defect caused by the primary surgery [[Figure 2] and Video 1].
The anterior half of the superior incision goes up along the anterior edge of the septal defect as high as possible [Figure 2] and [Figure 3], and crosses over to the left side and proceeds forward anteriorly along the upper portion of the nasal septum on the left side [Figure 2] and [Figure 3], turning down on the free border of the septal cartilage to join the inferior incision [[Figure 2], [Figure 3] and Video 1]. The flap is harvested after placement of these incisions and placed onto the repair site [Figure 4].
In comparison with the standard Hadad flap, this technique differs only in the posterior half of the superior incision [Table 1]. It is preferable to place the incisions in the region of the choanae with a surgical scalpel rather than with cautery to prevent the chances of thermal injury to the vascular pedicle.
The lower portion of the right posterior nasal septum is raised [Figure 5] and transposed onto the left side, which is in continuity with the lower portion of the posterior left nasal septum [Figure 3] and [Figure 6]. Thus, the right posterior nasal septal mucosa becomes the upper portion of the posterior half of the NSF [Figure 6], and the resultant NSF has full thickness of bilateral posterior nasal mucosa incorporated in it. The entire procedure has been summarized in the Video 1.
In all the four cases [Table 2], there was continuity of mucosa from posterior nasal septum over the intact edge of the left sphenoid sinus floor, and it was noted to be extending laterally over the roof of the choana [Figure 1]. Therefore, the flap was raised on the left side, with the preservation of the left sphenopalatine pedicle because the area between the left choanal roof and left sphenoid sinus opening edge was kept intact. The NSF was used for ESBR in all patients and postoperative MRI showed contrast uptake in the NSF, indicating adequate perfusion within the flap [Figure 7]. This is significant because it confirms the presence of a functioning vascular pedicle within the NSF.
All patients had successful ESBR in this series because none of them developed postoperative CSF leak. Nasal block due to crusts was noted postoperatively, which was treated with endoscopic cleaning and saline nasal spray. Superficial fungal colonization was noted on the crust over the NSF in one patient, which was removed. The NSF was noted to be completely integrated with the rest of the mucosa [Figure 8] in another patient who reported for post operative review after 6 months.
The NSF ,,,,, remains the vascular flap of choice in ESBR. However, the presence of a septal perforation, bilateral large sphenoidotomies, and compromised sphenopalatine pedicles due to cauterization, postoperative scarring, and mucosal adhesions can make the raising of the NSF challenging in revision cases. Patel et al., described the placement of other vascular flaps such as the inferior turbinate flap, middle turbinate flap, lateral nasal wall flap, temporoparietel flap, pericranial flap, buccinator flap, palatal flap, and occipital flap as a solution for this scenario. Even though these flaps give goods results, whenever possible, the NSF should be raised in view of its vascularity, ability to cover a large area of the skull base, and the excellent arc of rotation, which make the placement of the flap easier.
Another clinical scenario where NSF becomes possible in revision surgery is by using the rescue flap technique  during the primary surgery. The cases we have selected underwent the primary surgery without the placement of the rescue incision, and therefore, our technique enables the surgeon to raise the NSF in the absence of rescue incision. As described by others, we have successfully taken down previously applied NSF in revision surgery in a few cases and replaced them successfully, with good results.
Brunworth et al., placed incisions inferior to the septal perforation and on the nasal floor and have raised the NSF in revision cases. In their cases, the mean width of the pedicle between the sphenoidotomy and roof of choana was 1.43 cm, and the mean distance between the inferior edge of the septal perforation and nasal floor was 1.07 cm. The minimum dimensions required for the pedicle to be viable is not yet determined; however, they recommend the endoscopic assessment of the mucoperiosteal tissue between the anterior septum and the sphenoid rostrum to check the viability of the NSF. Audible signals on acoustic Doppler sonography noted on the flap pedicle were considered to be a good indicator of vascular viability prior to harvesting of the flap in revision cases.
Contrary to Brunworth's technique, we prefer not to place the superior incision near the septal perforation for fear of injuring the remaining vascular pedicle. Instead, the superior incision starts on the left side of the previous sphenoidotomy, crosses the midline onto the right side, and then courses along the right side of the septum. The inclusion of bilateral full thickness posterior septal mucosa in the NSF [Figure 6] minimizes the risk of injury to the remnant vascular pedicle as the incisions are placed away from the potential area of the vascular pedicle. The net result is a flap with a larger surface area posteriorly, which ensures good vascularity. Crusting at the donor area of the septum can be reduced by placing free mucosal grafts from the middle turbinate  should the latter have been removed for better access during the surgery. The reverse flap  is not an option in our technique as bilateral posterior nasal septal mucosal tissues are included the NSF.
The presence of an intact mucoperiosteum between the septum and any one of the roofs of the choana, resting on an intact sphenoid sinus floor edge, as judged on an endoscopic and radiological assessment [Figure 1], may be considered as sufficient evidence for the presence of a viable pedicle in the NSF. The serial MRI scans at regular intervals for postoperative follow-up will further confirm the presence of a functioning vascular pedicle within the NSF. The absence of the bony edge of the sphenoid sinus floor [Figure 1] may be considered as a relative contraindication for this technique because the raising and subsequent rotation of the NSF become difficult in this scenario.
This is a useful technique for the reconstruction of the sellar defect in revision surgery where a vascular flap is required. While it is true that closure of the CSF leak may be done with grafts, the vascular flap is known to give better results. In a revision case, one needs to avoid CSF leak recurrence, and hence, it would be prudent to use a vascular flap for the skull base defect repair. Another advantage of our technique is that it offers the option of harvesting the NSF in the absence of prior rescue incision placement and when there is no prior NSF for takedown and repositioning.
The broad-based pedicle with bilateral full thickness posterior nasal septal mucosa in it forms the basis for assuring the adequacy of the remnant vascular pedicle in the NSF raised in this technique. Thus, the need to measure the surface area of the pedicle  and the usage of acoustic Doppler assessment of the pedicle  are avoided in this technique.
This technique has the limitation of being possible where only posterior septostomy has been done previously and one sphenopalatine pedicle is noted to be intact on the endoscopic and radiological assessment. Another disadvantage is that it involves bilateral flap elevation, even though this is limited to the posterior part of the septum. This needs extra time and careful flap elevation to avoid flap tear because the tissues may be adherent due to posterior operative scarring. In comparison with the other intranasal vascular flaps, which are the inferior turbinate flap, middle turbinate flap, and lateral nasal wall flap [Table 3], we still feel the usage of the revision NSF is the better option. While the technique does not individually identify the vascular pedicle with the Doppler ultrasound, the postoperative scans showed significant vascularity in this NSF, which had resulted in the good outcome achieved [Figure 7] and [Figure 8].
In revision EEA, a vascular flap is desirable to get better ESBR results. Though the number of cases in this study is small, this paper highlights a new operative technique, wherein one can harvest the NSF in revision cases. The presence of at least one intact sphenopalatine pedicle, as noted in the endoscopic and radiological assessment, is necessary for this technique. The need for measuring the surface area of the pedicle and the acoustic Doppler assessment of the pedicle are avoided in this method.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7], [Figure 8]
[Table 1], [Table 2], [Table 3]