Endoscopic endonasal repair of sphenoid cerebrospinal fluid leak, review and institutional experience

INTRODUCTION

Cerebrospinal fluid leak happens because of communication between the sterile subarachnoid space and the outer environment.

Repair of CSF leaks earns its importance due to the possibility of being complicated with intracranial infections and become life threatening if left untreated.

Initially trans-cranial approach to the skull base was used, however this approach is very aggressive and associated with high morbidity and mortality. Thus, since the introduction of endoscopic sinus surgery this method has become the standard of care in surgical management for the rhinogenic CSF leaks with a success rate of 90% in first attempt and up to 97% in the second attempts.^1,2

It must be mentioned that managing CFS leaks originating from the Sphenoid sinus poses unique difficulties in approach and visualization due to their anatomical location. And they should be handled in great delicacy because of the anatomic proximity to vital structures, namely the carotid artery and the optic nerve.

Another difficulty is the placement of underlay graft in this area, which might be endangering the adjacent neuromuscular structures.

In this study we aim to present the cases with symptomatic sphenoid CSF leak and our institutional experience in managing such cases. Also to give a summery of the current knowledge, focusing on diagnostic strategies and management details, from the review of the literatures.

METHODS

A retrospective study was performed for all cases of sphenoid sinus CSF leak, that have been presented to KFSH-D between November 2003 to December 2017.

(This hospital is a tertiary and referring hospital covering the eastern provision with a population of more than 4 million.)

Inclusion criteria: All cases with symptomatic sphenoid CSF leak presented to KFSH-D, from November 2003 to December 2017. Both genders and all ages were included.

Exclusion criteria: CSF leaks from sinuses other than sphenoid, and cases with iatrogenic leak during skull base and pituitary surgeries that were repaired immediately in the same surgery were not included. Also cases of traumatic CSF leaks that were successfully managed with conservative methods were excluded.

A total of 12 cases were found to fulfill inclusion and exclusion criteria.

Chart review done with data point collection for the age, sex, presentation and aetiology, localization methods including HRCT, CT and MRI cysternography and intra-operative cysternography.

The operative notes were reviewed and the post-operative records and clinic visits were recorded and analyzed using SPSS software.

And the U.S. National Library of Medicine (PubMed) database was searched for “Sphenoid CSF Leak”. And the citations of the retrieved manuscripts were searched to identify additional relevant articles.

RESULTS

We had a total of 12 cases, out of them 11 cases sought the help of otolaryngologist after an episode of meningitis. And the one that didn’t developed meningitis was refereed by neurosurgery department because of typical history of unilateral clear and spontaneous rhinorrhea with a brain tumour invading the bony sphenoid.

4 cases had a history of trauma. Road traffic accident in 3 of them and a gunshot injury in one of them was the traumatic mechanism (Table 1).

In our study all cases underwent plain high resolution CT (Figure 1) and intraoperative cysternography as the main localisation method. We had 4 cases with questionable sites of bony dehiscence on plain CT so they underwent CT cisternography for further evaluation. One of the cases had multiple leak sites, therefor navigation assisted CT (Figure 2) was used intraoperatively to detect all the sites of leakage.

MRI was used in 5 cases with suspected meningocele on the CT, and 4 of them had meningoceles (Figure 3).

The operative notes were reviewed and the post-operative records and clinic visits were recorded and analysed.

After induction of general anaesthesia, nasal cavity was decongested using xylometazoline soaked packing. By the help of neurosurgeons lumbar drain is placed and 0.2 mL fluorescein 10% is diluted in 10 ml CSF and slowly injected intrathecally (Figure 4).

Intraoperative fluorescein not only helps with localisation, but also helps insuring a watertight closure. Another method to use flurocein is by topical application and in this method there is no risk for central nervous system complications (Figure 5).

After that the patient is handed back to the otolaryngologist for the endoscopic repair. Starting with examination and identification of sphenoid osteum. The surgeon then proceeds to a wide sphenoidotomy.

For the sake of a better visualisation, it might be necessary to remove the supreme and superior turbinates or the posterior part of the middle turbinate. And in some cases we needed an ethmoidectomy to get better access for instrumentation.

If leak is located in the lateral recess the surgical field should be extended through transpterygopalatine or transmaxillary approaches.

Once the skull base defect is optimally exposed, herniated meningocele or meningoencephaloceles should be reduced into the intracranial cavity. That is easily achieved using a bipolar diathermy.

It should be mentioned that in the setting of an encephalocele, brain tissue that has herniated into the nasal cavity is rarely functional and is considered a potential intracranial source of infection if not resected.

A crucial step is that the sinonasal mucosa around the bony defect must be carefully removed to allow the graft to adhere securely to the bony skull base and prevent future mucocele formation at the surgical site.

We used mucoperiosteal flap, harvested from the nasal septum as the primary layer for all closures. It was secured in place using glue.

After applying the coverage flap, surgi-cell was used to avoid adherence of the flap to the nasal cavity walls. By using the sponge self-absorbable packing, we avoided the risk of removing the flap postoperatively during pack removal, that could happen due to adherence of the the pack to the flap by blood clot or glue.

The importance of cooperative anesthesiologists is not to be forgotten, to achieve a bloodless field and a smooth recovery from anesthesia. And cases been reported in which main cause of failure was suspected to be a stressful extubation.

We had two cases with recurrence. Both had increased ICP plus one had a large defect due to a gun shot wound. Both were managed in first time with a single on-lay method and after recurrence a more secure multilayer method plus a VP shunt was used, which was found successful after 2 and 5 years of follow up (Table 1).

Out of 6 cases of spontaneous CSF leak, half had increased ICP ant the other half had normal ICP.

The 3 cases that had an increased ICP needed more than simple one layer on-lay method. And when we used that method for one of them, the CSF leak recurred. Eventually, 2 of them ended up with VP shunt and the last one needed multilayer closure.

The 3 cases of traumatic RTA leaks and 2 out of 3 cases of spontaneous leaks without increased ICP were managed successfully with simple one layer unlay closure. The last case of spontaneous leak without increased ICP had a large defect and needed multilayer closure to insure no recurrence. In cases where mucoperiosteal and Haddad flaps were taken from the nasal septum, we found no donor site complications. Re-epithelization happened after around 2 weeks. There were no cases of nasal perforation.

Majority of our cases stopped follow up clinic visits after 2 or 3 years of improvement. There were no significant complications with the cases other than recurrence in 2 cases (Table 1).

DISCUSSION

Cerebrospinal fluid leak can only occur when all the layers separating the CSF-containing subarachnoid space from the nasal cavity are breached. This includes the arachnoid and dura matters plus a bony defect and nasal mucosa. However this is only a prerequisite and for the active flow to happen there must be a pressure gradient exceeding the tensile strength of the mucosal surface.³

The management of rhinogenic CSF leak has developed tremendously since the first published successful repair done by Dandy in 1926.⁴ Dandy described a transcranial reconstruction with placement of a facia lata graft. With the heavy burden of a full craniotomy and its morbidity, plus the high recurrence rate reaching up to 27%, this approach was soon out of favour.

In 1948 Dohlman described a naso-orbital incision, which was the beginning of the extra cranial approaches.⁵

As all the other rhinologic surgeries the CSF leak repair underwent a revolution with the introduction of endoscopic techniques. Papay et al was the first to describe the rigid endoscopic repair in 1989.⁶ This minimally invasive method reported to have a success rate of up to 90% in first attempt and 97% in the second attempt.⁷ Due to its safety and the great success rate this approach has become the mainstay of treatment nowadays.⁸

To begin the management of a patient with CSF leak, we should be able to recognise the clinical presentation of such cases. The patient will typically complain of a unilateral, clear watery nasal discharge that is flowing spontaneously with a salty or metallic taste. Some other characteristic complaints may include a headache that improves with rhinorrhea, or a rhinorrhea that is increasing with leaning forwards or straining. The CSF leak should be also suspected with unexplained recurrent history of meningitis with or without such nasal discharges.

The workup of a patient with CSF leak should include; first the confirmation of the presence of CSF and second the localization of the lesion to guide the surgical repair in case needed.

Glucose oxidase strips can be used to test for the presence of CSF, due to its higher glucose level than the usual nasal secretions. However this test is a very rough estimate with a high rate of false positives, because of reducing substances in tears and nasal secretions in addition to the false negatives that may happen with bacterial meningitis. ^9-11

The more reliable chemical markers are beta-2 transferrin which is far more specific and Beta trace protein that is both specific and sensitive.^12,13

CSF tracers can be used with imaging studies to detect an active flow to the nasal cavity, which both helps in the confirmation of the leak and can help in localization. These studies include radionuclide cisternography, CT cisternography and MRI cisternography. The radionuclide cisternography needs an intrathecal injection of radioactive dye and has a very poor spatial localization.¹⁴ CT cisternography also needs an intrathecal injection of the dye but gives a much better localization specially for the bony defect that is important in surgical planning.¹⁵ MRI cisternography has the advantage of not needing intrathecal injection, but a relatively poor localization in compared to CT cisternography.¹⁶

In a review by Zapalac et al, beta 2 transferrin was suggested for the confirmation of presence of CSF while high resolution CT without cisternography was suggested as the diagnostic strategy.¹⁷ In this case MRI can be used for a better assessment of soft tissue structures including possible concurrent meningoencephalocele, or soft tissue masses.

One of the most used localization techniques is intraoperative intrathecal fluorescein injection while doing endoscopic examination. Although concerns have been made regarding its safety, the latest studies announced that low-dose fluorescein (50 mg or less) is unlikely to be associated with adverse events, because most complications seem to be dose related.¹⁸

After the proper diagnostic evaluation we can proceed to treatment strategies for the leak, which may be conservative or surgical.

The surgical approach to the sphenoid sinus can be challenging due to its location, and even with the use of angled endoscopes it might be difficult to visualize the lateral wall of the sinus.¹⁹ That is why if leak is located in the lateral recess the surgical field should be extended through transpterygopalatine approach.²⁰

The other challenge is the proximity of the sphenoid sinus to some neuromuscular structures, which may render the placement of an underlay graft impossible. In such instances, obliteration of the sinus is advised.^21,22

CONCLUSION

Endoscopic endonasal approach for repair of sphenoid CSF leak was found to be both effective and safe. We advise using high resolution CT and intraoperative fluroscein for a reliable localization. And we found no complications with use of correct concentration of fluroscein.

Funding: No funding sources

Conflict of interest: None declared

Ethical approval: The study was approved by the Institutional Ethics Committee

REFERENCES

1. 1. Kim E, Russell PT. Prevention and management of skull base injury. Otolaryngol Clin North Am. 2010;43(4):809–16.
2. 2. Bernal-Sprekelsen M, Alobid I, Mullol J, Trobat F, Tomás-Barberán M. Closure of cerebrospinal fluid leaks prevents ascending bacterial meningitis. Rhinology. 2005;43(4):277–81.
3. 3. Bhalodiya NH, Joseph ST. Cerebrospinal fluid rhinorrhea: endoscopic repair based on a combined diagnostic approach. Indian J Otolaryngol Head Neck Surg. 2009;61(2):120-6.
4. 4. Engelhardt P, Trostdorf E. Dandy Walker syndrome with air Pneumocephalus. Arch Surg. 1926;12:949-82.
5. 5. Dohlman G. Spontaneous cerebrospinal rhinorrhea. Acta Otoiaryngol SuppI. 1948:67:20-3.
6. 6. Papay FA, Benninger MS, Levine HL, Lavertu P. Trans- na.sal transsepial endoscopic repair of sphenoidal cerebral spi-nal fluid fistula. Otoiaryngol Head Neck Surg. 1989;101:595-7.
7. 7. Hegazy HM, Carrau RL, Snyderman CH, Kassam A, Zweig J. Transnasal endoscopic repair of cerebrospinal fluid rhinorrhea: a meta-analysis. Laryngoscope. 2000;110:1166–72.
8. 8. Senior BA, Jafri K, Benninger M. Safety and efficacy of endoscopic repair of CSF leaks and encephaloceles: a survey of the members of the American Rhinologic Society. Am J Rhinol. 2001;15:21–5.
9. 9. Kirsch AP. Diagnosis of cerebrospinal fluid rhinorrhea: lack of specificity of the glucose oxidase test tape. J Pediatr. 1967;71:718–9.
10. 10. Katz RT, Kaplan PE. Glucose oxidase sticks and cerebrospinal fluid rhinorrhea. Arch Phys Med Rehabil. 1985;66:391–3.
11. 11. Calcaterra TC. Diagnosis and management of ethmoid cerebrospi-nal fluid rhinorrhea. Otolaryngol Clin North Am. 1985;18:99–105.
12. 12. Nandapalan V, Watson ID, Swift AC. Beta-2-transferrin and cerebrospinal fluid rhinorrhoea. Clin Otolaryngol. 1996;21:259–64.
13. 13. Arrer E, Meco C, Oberascher G, Piotrowski W, Albegger K, Patsch W. beta-Trace protein as a marker for cerebrospinal fluid rhinorrhea. Clin Chem. 2002;48:939–41.
14. 14. Yamamoto Y, Kunishio K, Sunami N, Yamamoto Y, Satoh T, Suga M, et al. Identification of CSF fistulas by radionuclide counting. AJNR Am J Neuroradiol. 1990;11:823–6.
15. 15. Colquhoun IR. CT cisternography in the investigation of cerebro-spinal fluid rhinorrhoea. Clin Radiol. 1993;47:403–8.
16. 16. Sillers MJ, Morgan CE, el-Gammal T. Magnetic resonance cisternography and thin coronal computerized tomography in the evaluation of cerebrospinal fluid rhinorrhea. Am J Rhinol. 1997;11:387–92.
17. 17. Zapalac JS, Marple BF, Schwade ND. Skull base cerebrospinal fluid fistulas: a comprehensive diagnostic algorithm. Otolaryngol Head Neck Surg. 2002;126:669–76.
18. 18. Keerl R, Weber RK, Draf W, Wienke A, Schaefer SD. Use of sodium fluorescein solution for detection of cerebrospinal fluid fistulas: an analysis of 420 administrations and reported complications in Europe and the United States. Laryngoscope. 2004;114:266–72.
19. 19. Tosun F, Carrau RL, Snyderman CH, Kassam A, Celin S, Schaitkin B. Endonasal Endoscopic Repair of Cerebrospinal Fluid Leaks of the Sphenoid Sinus. Arch Otolaryngol Head Neck Surg. 2003;129(5):576-80.
20. 20. Al-Nashar IS, Carrau RL, Herrera A, Snyderman CH. Endoscopic transnasal transpterygopalatine fossa approach to the lateral recess of the sphenoid sinus. Laryngoscope. 2004;114(3):528-32.
21. 21. Mao VH, Keane WM, Atkins JP, Spiegel JR, Willcox TO, Rosen MR, et al. Endoscopic repair of cerebrospinal fluid rhinorrhea. Otolaryngol Head Neck Surg. 2000;122(1):56-60.
22. 22. Buchman CA, Civantos FJ, Casiano RR. Skull base cerebrospinal fluid leaks and encephaloceles, in Advances. Otolaryngol Head Neck Surg. 1999;13.