Endoscopic Reconstruction of the Sellar Floor by Extended Inferior Turbinate Flap in Recurrent Pituitary Tumors

Abstract

Objective  This aim of this study was to address the outcome of endoscopic reconstruction of the sellar floor by extended inferior turbinate flap.

Patients and Methods  This is a retrospective study of 34 patients with a recurrent pituitary tumor. They were treated between March 2018 and December 2021 by endoscopic extended endonasal approach with the reconstruction of the sellar floor by an extended posterior pedicle inferior turbinate flap. The clinical and radiological follow-up was performed immediately postoperation and regularly every 3 months up to 1 year, and the available data from the last follow-up visit were included in the analysis.

Results  The patients' age ranged between 40 and 65 years, with a slight female predominance (55.9%). Headache was the main presentation (47.1%), and functional tumors were found in 50.0% patients. Visual disturbances were field defects among 61.8% and papilledema among 52.9% patients. Preoperative endoscopy revealed postseptectomy as the significant finding (73.5%), followed by postseptectomy and adhesion (14.7%) and finally postseptectomy and hypertrophied inferior turbinate (11.8%). Total tumor resection was achieved in 76.5%, visual improvement was recorded in 52.9%, and no complications were reported in 82.4% patients. Cerebrospinal fluid (CSF) leak was not reported in any of the studied patients. Finally, total resection was significantly associated with younger age, non-functioning tumor and improvement of headache.

Conclusion  The extended inferior turbinate flap is an effective and safe approach for sellar floor reconstruction in endoscopic endonasal surgery for recurrent pituitary tumors. The extension overcomes the relatively small inferior flap and its limited arc of rotation.

Introduction

Recurrent pituitary macroadenomas represent a high long-term recurrence rate (∼7–12% at 10 years), even after complete surgical resection. Endoscopic surgery is the primary treatment that causes symptoms related to mass effects on critical neurovascular structures or endocrinal manifestations. The expanded endonasal approach (EEA) provides a good exposure to the skull base and intradural pathology while reducing the morbidity associated with traditional craniofacial approaches. ^{1 2 3}

Reconstruction of anterior skull base defects resulting after EEA is more challenging to reconstruct due to their huge size defect. Failure to achieve adequate separation between the intracranial partition and sinonasal tract can lead to complications (e.g., meningitis, cerebrospinal fluid [CSF] leak, and pneumocephalus). The primary reconstruction concepts are to detach the cranial cavity from the sinonasal tract and safeguard vital neurovascular structures. A secure separation safeguards against postoperative CSF leaks, meningitis, vascular blowouts, and pseudoaneurysms secondary to desiccation or primary vessel infection. ^{4 5}

Reconstruction of the skull defect can be achieved using various grafting techniques, either by free tissue flap or by local vascularized flap. The free tissue flap has several disadvantages, such as its unsuitability for significant defects; delayed healing, which could lead to flap detachment; and the need for another incision in the donor site. A vascularized flap provides a reliable reconstruction of significant skull base defects with low rates of postoperative CSF leak when compared with free tissue grafts. ^{6 7 8 9}

Local vascularized flaps gained wide acceptance and became the principal reconstructive choice due to their advantages (e.g., ease of elevation, low morbidity of the donor site, low rate of complications, and the tendency for rapid and complete healing). The nasoseptal flap (NSF) has become the mainstay of endonasal reconstruction as it is versatile, has a good arc of rotation, and provides a large surface area. However, in cases of previous posterior septectomy or prior wide sphenoidotomies, an alternative flap should raise the attention for another technique of reconstruction. ^{10 11}

Extended inferior turbinate flap (EITF) is developed on the basis of the inferior turbinate artery (ITA), a terminal branch of the posterior lateral nasal artery (LNA) that arises from the sphenopalatine artery (SPA). It supplies most of the lateral nasal wall, including the inferior and middle turbinates. The posterior pedicle inferior turbinate flap (PPITF) permits a reduction of the healing time due to rapid mucosalization, especially after EEA in recurrent pituitary macroadenoma. ¹² Here, we present our unique experience with posterior EITF for the management of skull base defects in recurrent pituitary tumors.

Patients and Methods

A retrospective study was performed on 34 patients with recurrent pituitary tumors from March 2018 to December 2021. All the patients were operated on by endoscopic extended endonasal approach (EEA) with the reconstruction of the anterior skull base by extended PPITF. All surgeries were completed at Al-Azhar University Hospitals (Damietta, Al-Hussein, and Sayed Galal University Hospitals; Otolaryngology and Neurosurgery departments).

The inclusion criteria were patients aged 40 to 65 years and patients with recurrent pituitary tumors in which a previous posterior septectomy was performed (absence of NSF).

Preoperative assessment of all patients was composed of a complete history taking and a detailed neurological and rhinological examination (e.g., visual field and nasal endoscopic assessment). Each patient had a plain X-ray, a computed tomography (CT) scan of the paranasal sinus, magnetic resonance imaging (MRI) of the brain with contrast, and full pituitary hormonal assessment.

Surgical Technique

The surgical procedure was done with Karl Storz Spice, a high-definition endoscopic camera, under general anesthesia in the supine position with careful head fixation and flexion of 30 degrees. A standard surgical preparation and draping were performed. A cotton pledge soaked with thick adrenaline was applied on and under the inferior turbinate, and then 1% lidocaine with 1/100,000 epinephrine was infiltrated into the inferior turbinate, nasal floor, and inferior meatus. The inferior turbinate was gently medialized to open the inferior meatus and then subsequently laterally fractured to gain access to the lateral nasal wall. The sphenopalatine foramen is identified posterior to the basal lamellae of the middle turbinate. The pedicle blood supply to the inferior turbinate can sometimes be visualized as pulsating, which aids in incision planning.

Incisions were performed with monopolar cautery. The superior incision begins anterior to the sphenopalatine foramen and continues anteriorly in a horizontal plane over the attachment of the inferior turbinate on the lateral nasal wall ( Fig. 1 ). The posterior incision ( Fig. 2 ) in the coronal plane starts posterior to the sphenopalatine foramen and descends vertically anterior to the eustachian tube, down to the nasal floor. This was then brought medially with extension to the nasal floor even up to the nasal septum, then brought anteriorly across the anterior end of the inferior turbinate ( Fig. 3 ). The two incisions were connected by a vertical incision at the head of the inferior turbinate ( Fig. 4 ). This incision was an S shape, starting from the superior incision, sloping around the contour of the head of the inferior turbinate, and into the inferior meatus. Care was provided to avoid disruption of the Hasner valve.

Fig. 1.

The superior incision.

Fig. 2.

The posterior incision.

Fig. 3.

Anterior vertical excision.

Fig. 4.

Connecting the incisions.

Careful elevation with a suction elevator instrument will help ensure flap viability ( Fig. 5 ). The bone of the inferior turbinate was left in place to remucosalize, and therefore minimizing the morbidity of the procedure. The flap was then tucked in the nasopharynx ( Fig. 6 ) and brought back up at the end of the case as an overlay of the defect. The flap is smooth in its regular rotation, so the mucosal side faces externally, and the pedicle is not kinked. A multilayered reconstruction is usually accomplished with a bone ship from the inferior turbinate and then followed by the flap as an overlay ( Figs. 7 8 9 10 ). The edges of the flap are covered with Surgicel, and the whole area is matted with fibrin glue. Saline-soaked Gelfoam W (Pfizer; New York City, NY, United States) follows. Doyle Silastic splints are sutured across the septum.

Fig. 5.

Flap elevation.

Fig. 6.

The flap was then tucked in the nasopharynx.

Fig. 7.

Elevation of the anterior edge.

Fig. 8.

Complete elevation of the flap.

Fig. 9.

Reposition of the flap.

Fig. 10.

Flap in position.

Follow-Up

The clinical and radiological follow-up was performed immediately postoperation and regularly every 3 months up to 1 year. The data of the last follow-up visit were included in the statistical analysis.

Ethical Considerations

The study protocol was approved by the institutional review board of Damietta Faculty of Medicine, and appropriate administration consent was obtained. A detailed operative report was written, and video documentation of the operation was obtained.

Data Analysis

Statistical analysis was done by the Statistical Package for the Social Science (SPSS) for Windows (Standard version 12; SPSS Inc, Chicago, IL, United States). The normality of data was first tested with a one-sample Kolmogorov–Smirnov test. Numerical variables were presented as a mean and standard deviation, while categorical variables were shown as frequency and percentage. Groups were compared by the independent sample's Student's “ t ” test, chi-squared test, or their equivalents according to the type of data. A p value < 0.05 was considered significant.

Results

The preoperative data of included patients are presented in Table 1 . The mean age was 50.88 years (range: 40–65 years), with a slight female-sex predominance (females represented 55.9%). Headache was the main complaint among 47.1%, and functional tumors were found in 50.0% (ACTH, growth hormone, and prolactin production were presented on 26.5, 11.8, and 11.8%, respectively). Visual field defects were reported by 61.8% of patients and papilledema by 52.9% of patients. Preoperative endoscopy revealed postseptectomy as the major finding (73.5%), followed by postseptectomy and adhesion (14.7%), and finally postseptectomy and hypertrophied inferior turbinate (HIT; 11.8%). Preoperative MRI showed that the lesion was suprasellar (26.5%), presellar (26.5%), suprasellar and parasellar (23.5%), parasellar (17.6%), and retrosellar (5.9%).

Table 1. Preoperative data of studied populations.

Variable		Statistics
Age (y)	Mean ± SD	50.88 ± 7.20
	Range (minimum–maximum)	40–65
Sex, n (%)	Male	15 (44.1)
	Female	19 (55.9)
Headache, n (%)		16 (47.1%)
Functioning tumor, n (%)		17 (50.0%)
Main function, n (%)	Nonfunctioning	17 (50.0)
	ACTH producing	9 (26.5)
	Growth hormone producing	4 (11.8)
	Prolactin producing	4 (11.8)
Papilledema, n (%)		18 (52.9%)
Visual field defects, n (%)		21 (61.8%)
Preoperative endoscopy, n (%)	Postseptectomy	25 (73.5)
	Postseptectomy and adhesion	5 (14.7)
	Postseptectomy and HIT	4 (11.8)
MRI, n (%)	Suprasellar	9 (26.5)
	Presellar	9 (26.5)
	Retrosellar	2 (5.9)
	Suprasellar and parasellar	8 (23.5)
	Parasellar	6 (17.6)

Abbreviations: ACTH, adrenocorticotropic hormone; HIT, hypertrophied inferior turbinate; MRI, magnetic resonance imaging; SD, standard deviation.

Operative and postoperative data are depicted in Table 2 . The operative time (hours) ranged between 1.4 and 2.8 hours (mean: 2 hours). Blood loss ranged between 100 and 300 mL (mean: 219.11 mL). There was total surgical resection in the majority of patients (76.5%) and subtotal resection was performed in 23.5% of patients. Visual improvement was recorded among 52.9% of patients, and no complications were reported among 82.4% of patients. However, complications that occurred were in the form of diabetes insipidus (14.7%) and acute coronary syndrome (2.9%). The CSF leak was not reported in any of the studied patients.

Table 2. Operative and postoperative data of studied populations.

Variable		Statistics
Operative time (h)	Mean ± SD	2.0 ± 0.39
	Range (minimum–maximum)	1.40–2.80
Blood loss (mL)	Mean ± SD	219.11 ± 55.06
	Range (minimum–maximum)	100–300
Surgical resection, n (%)	Total	26 (76.5)
	Subtotal	8 (23.5)
Visual improvement, n (%)	Yes	18 (52.9)
	No	3 (8.8)
	NA	13 (38.2)
Complications, n (%)	None	28 (82.4)
	Diabetes insipidus	5 (14.7)
	Acute coronary syndrome	1 (2.9)
CSF leak, n (%)	No	34 (100.0)
	Yes	0 (0.0)

Abbreviations: CSF, cerebrospinal fluid; SD, standard deviation.

Total resection was significantly associated with younger age, nonfunctioning tumor, and improvement of headache ( Table 3 ).

Table 3. Associated factors with total resection of the tumor.

Variable		Total resection ( n  = 26)	Subtotal resection ( n  = 8)	Test	p
Age (y)		49.46 ± 6.78	55.50 ± 6.96	2.18	0.036 *
Operative time (h)		2.02 ± 0.41	1.93 ± 0.32	0.53	0.59
Blood loss (mL)		221.15 ± 51.32	212.50 ± 69.43	0.38	0.70
Sex, n (%)	Male	11 (42.3)	4 (50.0)	0.14	0.70
	Female	15 (57.7)	4 (50.0)
Function, n (%)	Functioning	10 (38.5)	7 (87.5)	5.88	0.015 *
	Nonfunctioning	16 (61.5)	1 (12.5)
Headache, n (%)	Yes	9 (34.6)	7 (87.5)	6.86	0.014 *
	No	17 (65.4)	1 (12.5)
Fundus, n (%)	Papilledema	14 (53.8)	4 (50.0)	0.04	0.84
	Normal	12 (46.2)	4 (50.0)
Visual filed, n (%)	Defect	17 (65.4)	4 (50.0)	0.61	0.34
	None	9 (34.6)	4 (50.0)
Visual improvement, n (%)	Yes	15 (57.7)	3 (37.5)	1.01	0.60
	No	2 (7.7)	1 (12.5)
	NA	9 (34.6)	4 (50.0)
Complications, n (%)	None	23 (88.5)	5 (62.5)	4.50	0.11
	DI	3 (11.5)	2 (25.0)
	ACS	0 (0.0)	1 (12.5)

Abbreviations: ACS, acute coronary syndrome; DI, diabetes insipidus.

^*indicate statistical significance.

Discussion

Primary pituitary gland tumors constitute 10 to 15% of all intracranial lesions. The common pathologies include pituitary adenomas, craniopharyngiomas, and cysts of the Rathke cleft. ^{13 14} The endoscopic endonasal surgical approaches have become the primary surgical technique to address sellar tumors. This is due to their advantages of better visualization, permitting a panoramic view of tissues and their surrounding vascular structures, with improved safety and efficacy. However, tumor recurrence that needs revision surgery is reported. ^{15 16 17 18}

Reconstruction of the sellar floor is of crucial importance in primary endoscopic endonasal approaches addressing the pituitary tumors to manage or guard against CSF leak. It has become a mandatory and essential requirement in revision surgery. Reconstruction had been performed on grafts, flaps, and synthetic materials in single or multiple layers, with heterogenic results. The pedicled NSF, introduced in 2006, is considered the cornerstone in endoscopic skull base reconstruction. ^{19 20 21 22 23} The EITF provides an additional alternative for reconstruction when the viable NSF is unavailable (as in revision surgery). ¹²

In this study, we address our clinical experience for the endoscopic reconstruction of the sellar floor in recurrent pituitary tumors by the EITF. The mean operative time was 2 hours, with minimal blood loss (mean: 219.11 mL), and total resection was achieved among 76.5% of patients, which was associated with visual improvement among 52.9% patients. Complications were absent in 82.4% of patients (specifically, no CSF leak was reported). These results reflected the overall efficacy and safety of the procedure. The main advantages that could explain higher effectiveness are the nature of the flap as a vascularized one. Thus, it could be used to close larger defects. ²⁴ The same authors reported that endoscopic endonasal surgery had been revolutionized by the NSFs, permitting the closure of significant skull base defects and a higher success rate in the prevention of postoperative CSF leak. In addition, Scagnelli et al ²⁵ and Pérez-López et al ²⁶ confirmed that the use of NSF reduced postoperative CSF, leading to its lowest reported rates (0.0–2.9%). The current results for EITFs yielded no CSF leak.

Fortes et al ²⁷ initially reported using PPITFs for the skull. The results revealed complete healing of all flaps that covered the entire defect.

The use of inferior turbinate vascularized flaps is not a new concept. Anterior flaps were described for the closure of septal perforations and internal lining of nasal defects. These flaps are based on the branches from the angular artery, a branch from the facial artery. However, the main blood supply of the inferior turbinate comes from the ITA, a terminal branch from the LNA, which is a terminal branch from the SPA. ^{28 29} The ITA enters the turbinate posteriorly on its lateral surface. ³⁰ The PPITF development for skull base reconstruction had been based on this artery. It provides a reliable vascular supply and good arc of rotation with sufficient reach for most of the ventrocaudal skull base defects.

Choby et al ¹² reported that the ETIF had successfully prevented a postoperative CSF leak in revision surgery for a case series of five patients. The main reason for selecting the ETIF is the previous utilization of NSFs. It provided a large and versatile flap that adequately covered the defect. In addition, no increase in postoperative crustation was observed in any patient. The extension of the flap was achieved by medial movement of the inferior incision to the septum, thus doubling the surface area of the flap and overcoming the main disadvantage of the inferior turbinate flap (potential smaller flap and limited arc of rotation). Yip et al, ³¹ in a case series, support the use of posteroinferior turbinate as a viable alternative for skull base reconstruction in the absence of the NSFs.

Although other local flaps are available when NSFs are no longer available. These regional vascularized flaps included pericranial flap, palatal flap, and temporoparietal fascia flap, with good results. However, these flaps are associated with increased surgical morbidity (e.g., hematoma, flap necrosis, or infections). ³²

The main clinical presentations of our patients were headache, visual disturbances, or hormonal disturbances. These are in line with previous findings in the literature. For example, Singh et al ³³ reported that diminished vision was the most common complaint (79%); acromegaly was reported in 21% of cases in their study. Their rate of total resection (64.0%) was lower than that in the present study (76.5%), and 71.0% of their patients had no postoperative complications (88.5% in the present study). In addition, transient diabetes insipidus was reported for 21% of patients (resolved within 3–4 days). Unlike the current work, one patient had a CSF leak managed by placing a lumbar drain. Visual improvements and reduced hormonal imbalance were reported in the present work.

To sum up, the value of the current study is based on the unique nature of treated patients (patients with recurrent pituitary tumors) and the extension of the inferior turbinate flap. The NSFs are the gold standard in managing skull base defects during endonasal transsphenoidal trans-sellar surgery for pituitary tumors. However, in revision surgery for current tumors, the NSFs are absent, and an alternative flap must be used. An inferior turbinate flap is a reasonable alternative, but it is relatively small with a limited arc of rotation. The extension of the flap overcomes these shortcomings.

There have been very few reported cases of the use of EITFs in the literature, most of which were performed on cadavers. Thus, the present study is valuable and includes a respectable number of patients. However, the sample size is still small to generalize the results of the study. The retrospective nature of the study is another limitation of the present study.