Trans-nasal endoscopic repair of cerebrospinal fluid rhinorrhoea: a retrospective analysis

INTRODUCTION

CSF rhinorrhoea is a rare but potentially devastating condition resulting from disruption of all the barriers which separate the subarachnoid space from the upper part of aero digestive tract which include arachnoid mater, duramater, skull base, mucosa of nasal cavity and paranasal sinuses.¹The resulting communication with CNS can result in multitude of infectious complications which can lead to significant morbidity and disastrous long term deficits for the patient. Main goal of treatment is to prevent ascending meningitis.² Etiologically, CSF rhinorrhoea is classified as traumatic or non-traumatic. Traumatic rhinorrhoea is further classified as accidental or iatrogenic. Accidental trauma is responsible for 90% cases of CSF rhinorrhoea of which 80% are due to motor vehicle accidents. Iatrogenic rhinorrhoea can follow transcranial endoscopic surgery via skull base.³ Few cases of CSF rhinorrhoea have been reported where neoplastic lesions are the causative factors.⁴ Presenting symptoms in CSF rhinorrhoea vary from clear nasal discharge and headache to changes in mental status. Most acute post-traumatic cases resolve with conservative treatment. In the event of chronic or prolonged leaks, surgical intervention becomes mandatory.⁵ While considering surgery, it becomes important to identify the etiology of leak, its location and size, shape of bony and meningeal defects and presence of concomitant intra-cranial hypertension or meningo-encephalitis.^3,5 Since these patients stand at high risk of complications such as brain abscess, meningitis or pneumo-encephalitis, therefore all persisting CSF leak cases should be repaired.⁶ Purpose of the present study is to review the management and outcomes of various cases of CSF rhinorrhoea in a tertiary care hospital.

METHODS

A retrospective study was conducted on the basis of medical records of 08 patients (6 males and 2 females) who had CSF rhinorrhoea and underwent trans-nasal endoscopic repair in Department of Otorhinolaryngology, Christian Medical College and Hospital, Ludhiana, Punjab (India) during the period from August 2016 to September 2017. Patients were assessed for demographic profile, etiology of CSF leak and its duration, endoscopic findings, radiographic findings, location of leak, surgical repair and post-operative outcome. Retrieved data was represented in frequency and proportion in tabulated form for evaluation of results.

Inclusion criteria

Inclusion criteria were cases with defect size less than 1.5 cms; cases involving cribriform, ethmoid, sphenoid or frontal recess region of skull base.

Exclusion criteria

Exclusion criteria were CSF leaks with duration less than two weeks; defect size more than 1.5 cms; cases involving posterior table of frontal sinus; leaks associated with multiple skull base fractures; leaks associated with tumour.

Procedure

All surgical procedures were done under general anesthesia, with patient lying supine and head end elevated 30 degrees. Left thigh was prepared to harvest fat and fascia graft when required. Nasal decongestion was achieved with 4% xylocaine with adrenaline 1:30000 packs. Middle turbinate was lateralized or resected to gain access to the cribriform area in case of leaks from the cribriform area. In case of defect in the fovea ethmoidalis, complete ethmoidectomy was done. Site of leak was visualized and confirmed by Valsalva manouvere. Nasal mucosa around the site of leak was partially removed. Meningocele/meningoencephalocele was cauterized at the stump and removed. Fat plug was placed in all cases by bath plug method and placed in underlay manner. Further, layered reconstruction was done with a combination of cartilage, fibrin glue, septal bone, middle turbinate mucosa, septal mucosa or fibrin glue. Bilateral nasal packs were placed which were removed on 3^rd postoperative day. Lumbar drain was placed in cases with larger defects and/or increased intracranial pressure. Patients were put on acetazolamide 250 mg 8 hourly for 1 week postoperatively.

RESULTS

In present study 08 patients (6 males and 2 females) were assessed. Mean age of patients was 28.50 years with range of 08-50 years (Table 1). Etiological factor for CSF leak was found to be accidental trauma in 63% cases. In one patient cause was iatrogenic resulting from polypectomy. Spontaneous rhinorrhoea was observed in 25% subjects. Majority of patients (75%) presented with intermittent leaks whereas 25% were suffering from continuous type of leakage. Thirty eight percent of cases had 1-2 episodes of meningitis (Table 2).

Radiographic examination revealed bony defects in all these cases with concurrent meningocele and meningoencephalocele in 13% and 50% cases respectively. Most common location of defect in this cohort was cribriform plate of ethmoid (75%) whereas it was fovea ethmoidalis in 25% patients. Cysternography was carried out in 25% patients to delineate the defect because of doubts in defining the lesion by HRCT/MRI in these cases (Table 3). All leaks were repaired with an underlay multilayered technique to restore the original anatomy layer by layer. Fat, middle turbinate (MT) mucosa and septal bone were used as graft to fill the defects in 50% cases. 13% of patients were treated by fat, MT mucosa and inferior turbinate mucosa. Fat, MT mucosa and septal cartilage were used in 13% and in 25% cases fat, septal mucosa and fibrin glue were used as grafts to seal the dehiscence. Post-operatively, lumbar drain was used in 38% cases only. No post-surgical complications were encountered in this study. Hospital stay in most of the cases (75%) was less than 2 weeks and in 25% of patients it was up to 4 weeks (Table 4). Mean follow up period was 12 months.

*MT: Middle turbinate

DISCUSSION

Extracranial approach for CSF leak was initiated by Dohlman in 1948 using naso-frontal incision which offered success rate of 60-80% with considerable reduction in risk of complications.^4,7 It was not until 1981 that Wigand described endoscopic repair of CSF rhinorrhoea which led to success rate of 90%.⁴ Since then this technique has gained popularity. It has been suggested that endoscopic repair be limited to gaps less than 1.5 cms in size but other authors have found no correlation between the success rate and size of the defect.⁶ Recent studies have reported success rate of 93% at first and 100% at second surgery.⁸ In present study, mean age of patients was found to be 28.50 years which is lower than the results presented by some authors.³ Lower mean age is related to the fact that majority of patients had traumatic leaks resulting from road traffic accidents and assaults which tend to involve the younger demographic.⁸ Preponderance of males over females has been observed in the present series. Traumatic leaks outnumber the non-traumatic leakages similar to the results presented by Safavi et al.³ An important aspect of success in such cases is identification of the exact site of CSF leak.^5,9 HRCT plays an essential role in identifying even the smallest defect and provides important information about the anatomical variations and guiding the surgeon in planning surgical repairs. MRI scanning was done to detect co-existing meningocele or meningoencephalocele in cases where causative factor was trauma. In present study 50% patients had meningoencephalocele and 13% had meningocele. These figures are higher than those observed by Ismail et al.⁶ Cysternography was done only in those cases where there was doubt in defining defect on CT/MRI scan. Commonest site of anatomical defect in majority (75%) of cases in our study was cribriform plate of ethmoid bone because of its inherent weakness due to presence of sieve like apertures for passage of olfactory nerve fibers. A variety of graft material was used ranging from fat, fascia lata, MT mucosa, septal cartilage or bone and fibrin glue depending upon the aetiology of leak, location and dimensions of bony dehiscence. Post-operative lumbar drain was used in cases that had larger defects and those with raised intra-cranial pressure. No post-operative complication was observed in any of patients in our study. These results are in concurrence with the earlier reports discussed in literature.^3,4,6,10 Success rate of 100% was recorded in this study similar to the results published by Ismail et al.⁶ Hospital stay was less than 2 weeks in majority of cases. Mean follow up period was 12 months with no recurrence in any of these cases.

CONCLUSION

Endo-nasal endoscopic closure is a safe and effective technique for repair of CSF rhinorrhoea. There is minimal intranasal trauma. An overall rate of successful repair was 100%. No surgical complications were encountered in present series. Etiology of leak and size of defect did not have any adverse effect on surgical outcome.

ACKNOWLEDGEMENTS

Authors duly acknowledge the help provided by the staff of medical record office, Christian Medical College and Hospital, Ludhiana for providing records related to the present study.

Funding: No funding sources

Conflict of interest: None declared

Ethical approval: Not required

REFERENCES

1. 1. Gilat H, Rappaport Z, Yaniv E. Endoscopic transnasal cerebrospinal fluid repair: A 10 years’ experience. Isr Med Assoc J. 2011;13:597-600.
2. 2. Daele JJ, Goffart Y, Machiels S. Traumatic, iatrogenic and spontaneous cerebrospinal fluid (CSF) leak: Endoscopic repair. B-ENT. 2011;7(17):47-60.
3. 3. Safavi A, Safavi AA, Jafari R. An empirical approach to the diagnosis and treatment of cerebrospinal fluid rhinorrhoea: An optimized method for developing countries. Malays J Med Sci. 2014;21(5):37-43.
4. 4. Presutti L, Mattioli F, Villari D, Marchioni D, Alicandri-Ciufelli M. Transnasal endoscopic treatment of cerebrospinal fluid leak: 17 years’ experience. Acta Otorhinolaryngol Ital. 2009;29(4):191-6.
5. 5. Mishra SK, Mathew GA, Paul RR, Asif SK, John M, Vaghese AM, et al. Endoscopic repair of CSF rhinorrhoea: An Institutional Experience. Iran J Otorhinolaryngol. 2016;28(84):39-43.
6. 6. Ismail AS, Costantino PD and Sen C. Transnasaltranssphenoidal endoscopic repair of CSF leakage using multilayer acellular dermis. Skull Base. 2007;17(2):125-32.
7. 7. Chappity P, Thakar A, Verma R. Endonasal endoscopic approach in management of paediatric CSF rhinorrhoea cases. Ind J Otolaryngol Head Neck Surg. 2015;67(1):88-92.
8. 8. Virk JS, Elmiyeh, Saleh HA. Endoscopic management of cerebrospinal fluid rhinorrhea: The Charing cross experience. J Neurol Surg B Skull Base. 2013;74(2):61-7.
9. 9. Lee DH, Lim SC, Joo YE. Treatment outcomes of endoscopic repairs of sinonasal cerebrospinal leak. J Craniofac Surg. 2011;22(4):1266-70.
10. 10. Husain M, Jha D, Vatsal DK, Husain N, Gupta RK. Neuroendoscopictrasnasal repair of cerebrospinal fluid rhinorrhoea. Skull Base. 2003;13(2):73-8.