Skip to main content
Journal of Neurological Surgery. Part B, Skull Base logoLink to Journal of Neurological Surgery. Part B, Skull Base
. 2022 Oct 12;84(6):585–590. doi: 10.1055/a-1934-9095

Superiorly Based Middle Turbinate Flap for Repair of Cerebrospinal Fluid Rhinorrhea: A Cadaveric Feasibility Study and Case Series

Mark B Chaskes 1,, Blair Barton 1, Michael Karsy 2, Mindy R Rabinowitz 2, Gurston G Nyquist 1, Christopher Farrell 2, James J Evans 2, Marc R Rosen 1
PMCID: PMC10581818  PMID: 37854540

Abstract

Objectives  Cerebrospinal fluid (CSF) leaks of the anterior cranial base are frequently repaired with endonasal, multilayered reconstructions. Vascularized tissue flaps are superior to free mucosal grafts and biomaterials in many cases. Limitations of previously described flaps include reach, rotation, pedicle availability, and postoperative sinonasal morbidity. The objective of this study is to describe the superiorly based middle turbinate flap, a novel vascularized mucosal reconstruction option, and to present a case series demonstrating flap utility.

Design  Cadaveric feasibility study with technical description and illustrative case series.

Setting  Tertiary medical center.

Participants  Three silicone-injected cadaveric specimens (6 sides); 7 patients with CSF rhinorrhea from bony dehiscence of the anterior cranial fossa repaired with a superiorly based middle turbinate flap.

Outcome Measures  Cadaveric feasibility, in vivo repair outcomes, sinonasal symptoms, and postoperative healing.

Results  Cadaveric dissection demonstrated a consistent vascular plexus arising from the anterior and posterior ethmoid arteries, originating at the superior attachment of the middle turbinate and traveling inferiorly to supply the mucosa of the middle turbinate. Mean surface area of the flap was 776.67 ± 114.60 mm 2 . The clinical series of 7 patients involved leaks around the cribriform plate and fovea ethmoidalis. There were no instances of repair failure. All cases showed rapid and complete remucosalization without significant sinonasal morbidity.

Conclusion  The superiorly based middle turbinate flap is a reliable, versatile, and effective option for a vascularized mucosal flap onlay that can be used in anterior skull base reconstruction. This flap is particularly useful in the repair of defects involving the cribriform plate and fovea ethmoidalis.

Keywords: skull base surgery, reconstruction, superiorly based middle turbinate flap, endoscopic skull base repair

Introduction

Anterior cranial base cerebrospinal fluid (CSF) rhinorrhea may result from spontaneous, iatrogenic, traumatic, or congenital causes. 1 Communication of the intracranial space with the nasal cavity and paranasal sinuses introduces the risk of retrograde intracranial infection and pneumocephalus, thus frequently necessitating surgical repair of the defect. 2 While the initial description of a purely endonasal approach in the treatment of CSF rhinorrhea in 1981 utilized an autologous free mucosal onlay graft alone, techniques have advanced such that reconstruction now often incorporates a multilayered repair. 2 3

While the materials and techniques employed in multilayered skull base reconstruction vary widely by institution and surgeon, many approaches incorporate vascularized mucosal flaps. 4 Vascularized flaps provide several distinct advantages over free mucosal grafts and biomaterials, including faster healing and less surface area required to remucosalize; and importantly, have proven more successful than free mucosal grafts in certain situations. 4 5 Several vascularized mucosal flaps have been described including the nasoseptal flap, the posterior pedicled inferior turbinate flap, and the middle turbinate flap. 6 7 8 Unfortunately, these flaps are not always available or appropriate in all situations.

We describe a novel, superiorly based middle turbinate flap supplied by a vascular plexus of the anterior and posterior ethmoid arteries. Due to the proximity of this flap to common sites of spontaneous and iatrogenic CSF leakage, and the unique blood supply of this flap, we find this flap to be a very useful adjunct in the multilayered repair of CSF leaks, particularly for those leaks originating at the cribriform plate and fovea ethmoidalis or when a nasoseptal flap is not available.

Methods

Three fresh, formalin-fixed cadaveric heads were acquired for dissection with silicone injection of arteries and veins (Science Care Inc, Phoenix, Arizona, United States). Flaps in cadavers were evaluated bilaterally and the surface area of the mobilized middle turbinate flap was measured by ruler.

Under direct endoscopic visualization, a vertical incision is made along the anterior head of the middle turbinate, extending from the superior attachment to the inferior most aspect of the head of the middle turbinate ( Fig. 1A ). A subperiosteal pocket is then elevated over the entire bony component of the turbinate using sharp and blunt dissectors ( Fig. 1B ). The bone of the middle turbinate is then removed in a piecemeal fashion. It is critical that the bony tunnel in which the middle turbinate branch of the sphenopalatine artery travels at the inferior most aspect of the turbinate is completely removed. This bony removal often requires extra attention to prevent accidental avulsion after removal of the superior bony support.

Fig. 1.

Fig. 1

Stepwise approach to superior based middle turbinate flap harvest. ( A ) A view of the right middle turbinate, with an opening incision marked out extending from the superior attachment down to the inferior most aspect of the head (dotted line). ( B ) Mucoperiosteal flaps are raised off of the bone of the middle turbinate bilaterally. ( C ) View from inside the middle turbinate with all of the bone removed. Superior releasing incisions are marked—a medially based flap (black dotted line) and a laterally based flap (white dotted line). ( D ) Posterior releasing incision is marked through the horizontal portion of the basal lamella (red dotted oval). ( E ) Transected posterior attachment of the middle turbinate with the middle turbinate branch of the sphenopalatine artery transected. ( F , G ) Unfurled medially based flap demonstrating the length of the flap extending to the bottom third of the inferior turbinate.

When the bony component of the middle turbinate has been removed completely, releasing incisions are made. The first releasing incision is made superiorly. Depending on the requirements of the vascularized flap, either the medial or lateral aspect of the middle turbinate mucosa can be released from the skull base superiorly ( Fig. 1C ). The releasing incision is then carried in a posterior and inferior trajectory, releasing the posterior attachment. During this step, the branches of the sphenopalatine artery supplying the middle turbinate are transected ( Fig. 1D and E ) and must be well cauterized. With all releasing incisions complete, the mucosa of the middle turbinate can be “unfurled” and rotated for repair of the anterior cranial fossa ( Fig. 1F and G ).

A review of patients was retrospectively performed after institutional review board approval. Relevant clinical factors were evaluated including the location of CSF leak and outcome at last follow-up. This technique was performed as part of a multilayered skull base reconstruction in seven patients for repair of CSF leaks originating at the cribriform plate and fovea ethmoidalis.

Results

Cadaver Dissection

Six middle turbinate flaps from three fresh cadaver specimens were raised. A vascular plexus, originating from the anterior and posterior ethmoid arteries, was consistently identified. This plexus, noted to originate at the superior aspect of the middle turbinate and travel inferiorly along the entire length of both the medial and lateral mucoperiosteal flaps of the middle turbinate, was consistently identified ( Figs. 1 and 2 ). All flaps were successfully designed and harvested using the technique described above. All flaps were of adequate size to suitably cover the entire ipsilateral cribriform and fovea ethmoidalis. The average surface area of the superior-based middle turbinate flap was 776.67 ± 114.70 mm 2 ( Table 1 ).

Fig. 2.

Fig. 2

Cadaveric dissection images. ( A ) Demonstration of the vascular plexus on both the medial and lateral mucosa of the middle turbinate, with a large branch of anterior ethmoid artery (AEA) noted. ( B ) The bony skull base has been resected demonstrating the AEA and vascular plexus. The superior releasing incision has been made on the lateral mucosal flap and the middle turbinate “unfurled.” A large branch of anterior ethmoid artery is noted supplying the plexus of the superiorly attached medial mucosal flap. ( C , D ) Vessels originating superiorly contributing to the plexus and extending the entire length of the mucosa. ( E ) Olfactory groove showing middle turbinate mucosa intact with submucosal vessels visible.

Table 1. Surface area of the six middle turbinates—measured ex vivo with the turbinate unfurled.

Length (mm) Height (mm) Surface area (mm 2 )
Middle turbinate #1 38 19 722
Middle turbinate #2 36 20 720
Middle turbinate #3 38 26 988
Middle turbinate #4 35 19 665
Middle turbinate #5 34 22 748
Middle turbinate #6 43 19 817

Clinical Series

This superiorly based middle turbinate flap was used in seven patients to reconstruct defects both medial and lateral to the middle turbinate ( Table 2 ). Five of these patients presented with CSF rhinorrhea secondary to an encephalocele from intracranial hypertension; one patient began leaking 3 weeks after endoscopic sinus surgery with identified borderline intracranial hypertension, and one patient was leaking from an iatrogenic defect that had been repaired 9 years prior.

Table 2. Seven-patient case series.

Age Sex Diagnosis Intraoperative opening pressure Site Vascularly based Postoperative leak Postoperative lumbar drainage (d) Complication
Patient 1 47 M Iatrogenic leak 10 Fovea ethmoidalis Medially No 1 Intracranial cyst
Patient 2 59 M Iatrogenic leak, IIH 18 Cribriform plate Laterally No 2 None
Patient 3 55 M IIH 23 Fovea ethmoidalis Medially No 3 Positional headaches from lumbar puncture
Patient 4 67 F IIH 21 Ethmoid/
posterior frontal sinus
Medially No 3 None
Patient 5 58 F IIH 22 Cribriform plate Laterally No 3 None
Patient 6 45 F IIH 22 Fovea ethmoidalis Medially No 1 None
Patient 7 44 F IIH 21 Fovea ethmoidalis Medially No 4 Transient anosmia

Abbreviations: F, female; IIH, idiopathic intracranial hypertension; M, male.

Defects were seen in the cribriform ( n  = 2), fovea ethmoidalis ( n  = 4), and junction of the ethmoid cavity and posterior table of the frontal sinus ( n  = 1) ( Fig. 3 ). Two superiorly based middle turbinate flaps were superolaterally based and rotated medially while the other five superiorly based middle turbinate flaps were superomedially based and rotated laterally. Five cases were repaired with a synthetic dural substitute inlay followed by the superiorly based middle turbinate flap onlay; one case was repaired with both a synthetic dural substitute inlay and onlay followed by the superiorly based middle turbinate flap onlay; and one case was repaired with a superiorly based middle turbinate flap alone. The middle turbinate flap was secured in place with dural sealant glue and absorbable packing material for graft stabilization in all cases. All patients had a lumbar drain placed intraoperatively and were drained from 1 to 4 days postoperatively at a rate of 5 to 10 mL/hour ( Table 2 ). There was no persistent CSF leak or flap failure at last follow-up in all patients. There was one case of development of a sterile intracranial cyst that was drained via on open approach on postoperative day 12. There was one case of self-reported anosmia that resolved by 2-month follow-up. All flaps showed complete remucosalization at 1-month follow-up ( Fig. 4 ).

Fig. 3.

Fig. 3

( A ) A preoperative computed tomography (CT) scan demonstrating an encephalocele of the fovea ethmoidalis. ( B ) A postoperative CT scan defect repair with a medially based superiorly based middle turbinate flap.

Fig. 4.

Fig. 4

( A ) Nasal endoscopy 5 months postoperatively demonstrating a well-healed laterally based middle turbinate flap for a cribriform defect. ( B ) Postoperative contrasted magnetic resonance imaging (MRI) demonstrating an enhancing flap (postoperative day 12).

Discussion

Study Findings and Benefit of Vascularized Flaps

Vascularized mucosal onlay flaps improve the success of multilayered cranial base reconstructions. 5 Here, we show the safety and efficacy of superiorly based middle turbinate flaps that can be used for repair of the cribriform plate and fovea ethmoidalis during the treatment of CSF rhinorrhea. Cadaver dissections are shown and were validated in a small clinical series.

The most commonly used endonasal vascularized flap is the nasoseptal flap. However, several morbidities—including the risk of postoperative septal perforation, saddle nose deformity, prolonged crusting, and olfactory disturbance—have been associated with the nasoseptal flap. 9 Further, the vascular pedicle to the nasoseptal flap is not always available either due to tumor involvement or due to previous surgery that has compromised the integrity of the vascular pedicle or its proximal blood supply. As such, the nasoseptal flap is not always the ideal choice in cranial base reconstruction. Alternative vascularized flaps have been described. These include intranasal flaps such as the posterior pedicled middle turbinate flap and the posterior pedicled inferior turbinate flap as well as extranasal flaps such as the temporoparietal fascia flap and the pericranial flap, which can be tunneled and introduced intranasally. 5

Advantages of the Middle Turbinate Flap

The initial description of the middle turbinate flap in 2009 relies on the middle turbinate branch of the sphenopalatine artery that enters the middle turbinate at its posterior inferior attachment. 8 While this flap has proven to be robust, with only one described flap failure in 51 described patients to date, there are limitations to the use of this posterior pedicled flap. 5 10 First, while the initial description of the posteriorly pedicled middle turbinate flap describes its utility in repairing defects of the fovea ethmoidalis, the dissection is admittedly difficult and its anterior reach may be limited. Second, by transecting the entire superior attachment of the middle turbinate, olfactory neuroepithelium is theoretically compromised. Additionally, it is pedicled off of a branch of the sphenopalatine artery, which may make this flap unavailable in some cases in which it would be of desirable use when the proximal vessel has been ligated.

The superiorly based middle turbinate flap avoids these potential limitations. First, the location of this flap makes it particularly versatile. Spontaneous and iatrogenic CSF leaks are most commonly located at the cribriform plate and fovea ethmoidalis. 11 While the superiorly based middle turbinate flap has a limited arc of rotation given the broad superior attachment, the location of the middle turbinate adjacent to the most common sites of spontaneous and iatrogenic CSF leaks make the superiorly based middle turbinate flap particularly well suited as a vascularized onlay option to reconstruct these defects. Without the need for significant flap rotation and tissue mobilization, the limited arc of rotation of the superiorly based middle turbinate flap is well suited for coverage of an ipsilateral anterior cranial base defect, and may reduce sinonasal symptoms postoperatively.

Further, the maintained superior attachment to the skull base during the bony removal portion of flap harvest provides sufficient counter tension to the downward dissecting force to allow for relative ease of dissection. In our experience, the time required to harvest the superiorly based middle turbinate flap is comparable to the time required to harvest other vascularized intranasal flaps.

In addition, it is described that middle turbinate mucosa contains olfactory neuroepithelium, particularly in the superior and posterior aspect. 12 13 While, to our knowledge, there are no papers that assess olfactory function specifically after a traditional pedicled middle turbinate flap is harvested, based on extrapolation from other studies, we hypothesize that olfaction may be compromised in some cases. Sowerby et al studied the impact of unilateral middle turbinate sacrifice on olfactory function and found 41% of patients experienced a decrease in olfactory function related to a unilateral middle turbinate sacrifice. 14 By preserving the mucosa at the superior attachment of the middle turbinate, particularly on the medial side, the superiorly based middle turbinate flap may more effectively preserve olfactory function ( Fig. 2E ).

Finally, and perhaps most importantly, our cadaveric dissection has proven a consistent vascular plexus supplying the mucosa of the middle turbinate that originates from the anterior ethmoid artery as branches exit the cribriform plate and enter the turbinate at its superior attachment. As other intranasal vascularized flaps—the nasoseptal flap, middle turbinate flap, and posterior pedicled inferior turbinate flap—are all based on branches of the sphenopalatine artery, the unique blood supply of this flap makes it particularly useful in cases when other intranasal flaps have a compromised blood supply.

Limitations

This study was a small sample feasibility and safety study. The use of cadaveric experiments aided in better understanding the utility of the middle turbinate flap. Future work can potentially evaluate the extent that such a flap can stretch depending on patient-specific anatomy. Additionally, objective data demonstrating pre- and postoperative olfactory function can be collected to better assess risk of olfactory disturbance. We acknowledge that six of seven patients included in this series were repaired with multilayered reconstructions, as is our standard institutional approach for the reconstruction of idiopathic and iatrogenic CSF rhinorrhea. However, given the robust vascularity of this flap, the superiorly based middle turbinate flap can be utilized as any other vascularized flap and can be employed within the context of other reconstruction techniques. Further study with additional patients can better evaluate the long-term efficacy and risks of this procedure.

Conclusion

The superiorly based middle turbinate flap is an effective option for a vascularized mucosal flap onlay for ipsilateral anterior cranial base defects. The ease of harvest and consistent anatomy make this a particularly reliable flap. Additionally, the ability of this flap to repair defects both medial and lateral to the superior attachment of the middle turbinate makes this flap versatile. The proximity of this flap to common sites of anterior cranial base CSF leakage, including the cribriform plate and fovea ethmoidalis, limit the required length of transposition and increase the possibility of preserving olfactory neuroepithelium during flap harvest. Finally, the unique blood supply to this flap makes it available when other commonly used intranasal flaps are compromised. The superiorly based middle turbinate flap is a reliable, versatile, effective, and low morbidity vascularized mucosal onlay option for anterior cranial base reconstruction.

Footnotes

Conflict of Interest None declared.

References

  • 1.Woodworth B A, Prince A, Chiu A G et al. Spontaneous CSF leaks: a paradigm for definitive repair and management of intracranial hypertension. Otolaryngol Head Neck Surg. 2008;138(06):715–720. doi: 10.1016/j.otohns.2008.02.010. [DOI] [PubMed] [Google Scholar]
  • 2.Lobo B C, Baumanis M M, Nelson R F. Surgical repair of spontaneous cerebrospinal fluid (CSF) leaks: a systematic review. Laryngoscope Investig Otolaryngol. 2017;2(05):215–224. doi: 10.1002/lio2.75. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Wigand M E. Transnasal ethmoidectomy under endoscopical control. Rhinology. 1981;19(01):7–15. [PubMed] [Google Scholar]
  • 4.Soudry E, Turner J H, Nayak J V, Hwang P H. Endoscopic reconstruction of surgically created skull base defects: a systematic review. Otolaryngol Head Neck Surg. 2014;150(05):730–738. doi: 10.1177/0194599814520685. [DOI] [PubMed] [Google Scholar]
  • 5.Gutierrez W R, Bennion D M, Walsh J E, Owen S R. Vascular pedicled flaps for skull base defect reconstruction. Laryngoscope Investig Otolaryngol. 2020;5(06):1029–1038. doi: 10.1002/lio2.471. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Hadad G, Bassagasteguy L, Carrau R L et al. A novel reconstructive technique after endoscopic expanded endonasal approaches: vascular pedicle nasoseptal flap. Laryngoscope. 2006;116(10):1882–1886. doi: 10.1097/01.mlg.0000234933.37779.e4. [DOI] [PubMed] [Google Scholar]
  • 7.Fortes F SG, Carrau R L, Snyderman C H et al. The posterior pedicle inferior turbinate flap: a new vascularized flap for skull base reconstruction. Laryngoscope. 2007;117(08):1329–1332. doi: 10.1097/mlg.0b013e318062111f. [DOI] [PubMed] [Google Scholar]
  • 8.Prevedello D M, Barges-Coll J, Fernandez-Miranda J C et al. Middle turbinate flap for skull base reconstruction: cadaveric feasibility study. Laryngoscope. 2009;119(11):2094–2098. doi: 10.1002/lary.20226. [DOI] [PubMed] [Google Scholar]
  • 9.Lavigne P, Faden D L, Wang E W, Snyderman C H. Complications of nasoseptal flap reconstruction: a systematic review. J Neurol Surg B Skull Base. 2018;79 04:S291–S299. doi: 10.1055/s-0038-1668158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Patel M R, Taylor R J, Hackman T G et al. Beyond the nasoseptal flap: outcomes and pearls with secondary flaps in endoscopic endonasal skull base reconstruction. Laryngoscope. 2014;124(04):846–852. doi: 10.1002/lary.24319. [DOI] [PubMed] [Google Scholar]
  • 11.Le C, Strong E B, Luu Q. Management of anterior skull base cerebrospinal fluid leaks. J Neurol Surg B Skull Base. 2016;77(05):404–411. doi: 10.1055/s-0036-1584229. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Pinna FdeR, Ctenas B, Weber R, Saldiva P H, Voegels R L. Olfactory neuroepithelium in the superior and middle turbinates: which is the optimal biopsy site? Int Arch Otorhinolaryngol. 2013;17(02):131–138. doi: 10.7162/S1809-97772013000200004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Griffiths C F, Barkhoudarian G, Cutler A et al. Analysis of olfaction after bilateral nasoseptal rescue flap transsphenoidal approach with olfactory mucosal preservation. Otolaryngol Head Neck Surg. 2019;161(05):881–889. doi: 10.1177/0194599819861340. [DOI] [PubMed] [Google Scholar]
  • 14.Sowerby L J, Gross M, Broad R, Wright E D. Olfactory and sinonasal outcomes in endoscopic transsphenoidal skull-base surgery. Int Forum Allergy Rhinol. 2013;3(03):217–220. doi: 10.1002/alr.21103. [DOI] [PubMed] [Google Scholar]

Articles from Journal of Neurological Surgery. Part B, Skull Base are provided here courtesy of Thieme Medical Publishers

RESOURCES