Eur Arch Otorhinolaryngol DOI 10.1007/s00405-015-3681-y
RHINOLOGY
Endoscopic endonasal multilayer repair of traumatic CSF rhinorrhea Ahmed Aly Ibrahim1 • Mohamed Okasha2 • Samy Elwany1
Received: 18 December 2014 / Accepted: 31 May 2015 Ó Springer-Verlag Berlin Heidelberg 2015
Abstract The incidence of traumatic CSF has increased in recent years due to increased incidence of road traffic accidents (RTA) as well the increasing number of endoscopic sinus surgeries (ESS). The objective of this study is to present our experience in management of traumatic CSF leaks using the endoscopic multilayer repair technique. Forty-two patients (aged 10–75 years, 30 males and 12 females) presenting with confirmed post-traumatic CSF rhinorrhea were operated upon between January 2007 and December 2013. The endoscopic multilayer technique was used in all cases. Electromagnetic navigation was used in some cases. All cases presented with intermittent watery rhinorrhea. The duration of the rhinorrhea ranged from 3 days to 1 year before repair. One case presented after 10 years from the causative trauma. Ten cases had a history of meningitis. Nine cases had more than one defect. Iatrogenic defects were larger than defects following accidental trauma. Two cases, following RTA, developed pseudoaneurysm of internal carotid artery. Ten cases had associated pneumocephalus. The mean duration of postoperative hospitalization was 6 days (range 4–8 days). The mean follow-up duration was 31.2 ?/- 11.4 months (range 16–48 months). None of our patient developed serious intra- or postoperative complications. Only one case required another surgery to repair a missed second defect. Post-traumatic CSF leaks can be successfully managed via the endonasal endoscopic route using the multilayer repair & Samy Elwany [email protected] 1
Department of Otolaryngology, Alexandria Medical School, Alexandria, Egypt
2
Department of Neurosurgery, Damanhur Medical National Institute, Damanhur, Egypt
technique. It is important to look for multiple defects in these cases. CT angiography is recommended for traumatic leaks involving the lateral wall of the sphenoid sinus to diagnose or exclude the development of pseudo-aneurysm of the internal carotid artery. Keywords CSF leak ESS RTA Skull base Endoscopic Repair
Introduction CSF rhinorrhea is an abnormal release of cerebrospinal fluid (CSF) from the subarachnoid space into the nasal cavity or paranasal sinuses. Approximately 80 % of posttraumatic CSF leaks result from nonsurgical (accidental) trauma. More than 50 % of the leaks are evident within the first 2 days, 70 % within the first week, and almost all present within the first 3 months [1]. The site of CSF leaks depends upon the type of trauma. The most common sites of CSF rhinorrhea following accidental trauma are the sphenoid sinus (30 %), frontal sinus (30 %), and ethmoid/cribriform area (23 %) [2]. On the other hand, The most common sites of CSF leak following endoscopic sinus surgery (ESS) are the roof of anterior ethmoid especially at the level of the cribriform plate (80 %), followed the frontal sinus (8 %) and sphenoid sinus (4 %). CSF leaks following neurosurgical procedures are usually located in the sphenoid sinus (67 %) because of the high number of pituitary tumors that are addressed via transsphenoidal approach [3]. The most common presenting symptom of CSF leaks is intermittent watery rhinorrhea that is classically positional in nature. In clinically suspicious cases, laboratory analysis of CSF markers will confirm the diagnosis [4]. Most cases
123
Eur Arch Otorhinolaryngol
of CSF Rhinorrhea following accidental trauma stop with conservative treatment within few weeks after trauma. Iatrogenic cases are much less likely to heal spontaneously. The surgical approach to CSF leak repair has undergone a major progression. Multiple technical advancements have contributed to this progression, but the most important amongst all has been the development and improvement of the endoscopic technique for identification, localization and repair of the leaking defects. The endoscopic endonasal approach has been reported to be well tolerated and much safer than external approaches. Moreover, it has several other advantages in comparison to external techniques. These advantages include accurate localization of the defect, preservation of olfaction in most cases, no brain retraction especially on the frontal lobe, less morbidity and less hospital stay. However, several authors have reported variable success rates [5–7] apparently due to differences in their diagnostic protocols, repair techniques, and other factors. In this paper we present our experience in endoscopic repair of 42 cases of post-traumatic CSF rhinorrhea. All repairs were performed using the endoscopic endonasal multilayer technique.
Materials and methods The study was conducted on 42 cases of post-traumatic CSF rhinorrhea who were referred to the Department of Otolaryngology—Head and Neck Surgery between January 2007 and December 2013. An informed consent from all patients was taken after they were fully informed about the surgical procedure and the use of their data in the research.
Preoperative evaluation and identification of the defect Careful history and complete general, otorhinolaryngological, and neurological examination were performed in all cases. Levels of beta-2 transferrin and/or beta-trace protein were measured in fluid samples collected from the nose in all cases. A cut-off point of 1.11 mg/ml for Betatrace protein was used to confirm the presence of CSF in the collected fluid [8]. Our routine radiologic protocol included multidetector computed tomography (MDCT) with ultra-thin 1 mm cuts (bone and soft tissue settings) and multiplanar reconstruction, and magnetic resonance (MRI) cisternography using high-resolution T2-weighted sequence. CT angiography was requested for fractures involving the lateral wall of sphenoid sinus to exclude the development of pseudoaneurysm of the internal carotid artery.
123
Operative technique We used Intrathecal fluorescein in all patients to further confirm CSF leak, localize the defect, and check the success of the repair at the end of surgery [9]. Intraoperative navigation was used in some patients (GE InstaTrackTM 3500 electromagnetic navigation system). 0°, 30° and 45° endoscopes were used during surgery. The procedure started with identification of the site of leak. Partial middle turbinectomy was done, in most cases, to create a wide surgical corridor and for use of its bone as a graft in multilayer repair. The mucoperiosteum around the defect was removed to prepare the area for grafting. A multilayer (sandwich grafting) reconstruction was performed using some or all of the following 5 layers according to the size, site, and etiology of the defect. 1st layer; fat plug harvested from the anterior aspect of the thigh (intracranial/intradural) 2nd layer; fascia lata tucked underneath the bony edge (intracranial/extradural). 3rd layer; bone harvested from the middle turbinate or nasal septum (intracranial/extradural). 4th layer; overlay graft over the defect using fascia lata (extracranial). 5th layer; free mucoperiosteal graft or vascularized nasoseptal flap or middle turbinate flap (extracranial). The underlay fascia lata graft was fashioned so that it had at least 1 mm of diameter extending circumferentially beneath the edges of the bony defect. Underlay grafting was not performed for defects in the cribriform plate. Other defects larger than 5 mm in diameter usually required the 5 layers of repair. Table 1 summarizes the tissues used for the repair. The grafts were stabilized in position with fibrin glue. Lack of fluorescein flow from the repaired defects was used to confirm achievement of watertight ‘gasket-seal’ at the end of the operation. The reconstruction area was further secured with SurgifloÒ, SurgicelÒ, and/or GelfoamÒ. The nose was then packed with MerocelÒ and the pack is left for 3–4 days postoperatively. In cases of sphenoid sinus defect the steps were modified, where exploration of anterior skull base was not mandatory. An angled 45° endoscope is used to visualize the lateral recess and confirm the presence of a CSF leak. In cases of post-hypophysectomy CSF leak following transsphenoidal surgery, repair was done aiming to close Table 1 Tissues used for CSF leakage repair (42 patients) Tissue used
No of patients (%)
Fascia lata
42 (100 %)
Fat plug
35 (83.3 %)
Bone graft
31 (73.8 %)
Nasoseptal flap
29 (69 %)
Middle turbinate flap
3 (7.14 %)
Eur Arch Otorhinolaryngol
the diaphragmatic arachnoid defect with absorbable material and to pack the sella and sphenoid sinus with fat patch rapped in SurgicelÒ supporting the repair in place. After packing of the sella, a layer of fascia lata was applied intrasellar in underlay fashion bellow the edges of the dura and another layer onlay covering the sella floor over the dural edges in the roof of the sphenoid sinus, then packing of the sphenoid sinus with fat. Nasoseptal flap was in all cases to cover the sellar defect. SurgicelÒ was applied to the edges of the nasoseptal flap then a layer of fibrin glue covered the flap. No intraoperative complications were encountered. Postoperative care included antibiotic prophylaxis, elevation of head of bed, food softeners, nasal packing (3–7 days) in all cases, and lumbar drainage in four cases. Hospital postoperative stay ranged from 2 to 7 days. The mean follow-up duration was 31.2 ± 11.4 months (range 16–48 months). Radiological follow-up (MRI) was done after 3 months and 1 year. The repair was considered
successful if there was no CSF leak at the end of the follow-up period.
Results The current study was carried out on 42 cases of traumatic CSF rhinorrhea. Thirty patients were males (71.4 %) and 12 females (28.6 %). Age varied from 10 to 75 years. Beta2 transferrin and/or beta-trace protein measurements confirmed the presence of CSF in the sample fluid of all patients. All cases presented with intermittent watery rhinorrhea. The duration of the rhinorrhea ranged from 3 days to 1 year before repair. One case presented after 10 years from the causative trauma (Fig. 1). 12 cases presented with headache, 8 cases with hyposmia/anosmia, and 10 cases with history of meningitis. Table 2 summarizes the etiology and some of the characteristics of the leaking defects. CSF leakage occurred
Fig. 1 A 36-year-old male with late post-traumatic CSF leak (10 years after trauma). (a and b) Coronal CT scans showing the defect in the posterior wall of right frontal sinus. (c and d) Coronal T2-weighted MRI showing frontal lobe encephalomalacia and leaking CSF into the right nasal cavity
Table 2 Etiology of CSF leakage and some characteristics of the leaking defect Etiology
Number of patients
Number of defects [1
1
Size of defect
Total
Mean site of defect (mm) CP/ASB
Pneumocephalus on CT
Sphenoid sinus
Accidental trauma
10
4
6
16
5.9
12
4
2
ESS
12
10
2
14
7.2
14
0
1
Hypophysectomy
15
15
0
15
12.3
0
15
4
5
4
1
6
9.5
6
0
3
Anterior skull base surgery
CP cribriform plate, ASB anterior skull base
123
Eur Arch Otorhinolaryngol Fig. 2 CSF leakage following endoscopic sinus surgery. a Coronal CT scan showing a big defect in the posterior cribriform plate and ethmoid roof (white arrowhead) with underlying fluid attenuation and in the right ethmoid and maxillary sinuses. b Coronal MRI showing bright CSF signal at the site of the leak (white arrowhead) as well as in the right ethmoid and maxillary sinuses
Fig. 3 Post-Hypophysectomy CSF leak. Sagittal a and Coronal b T2-weighted MRI showing bright CSF signal in the sella and sphenoid sinus (black asterisk). A frontal pneumocephalus is also seen (white asterisk) in the sagittal section
after RTA (7 cases), falling from height (3 cases), endoscopic sinus surgery (12 cases), hypophysectomy (15 cases), and skull base procedures (5 cases). More than one defect was seen in six out of 10 cases (60 %) following accidental trauma, two out of 12 cases (16.6 %) following endoscopic sinus surgery, and one out of five cases following anterior skull base surgery. Post-surgical defects (Fig. 2) tended to be larger than defects caused by accidental trauma. Pneumocephalus was detected in 10 cases (Fig. 3). In two cases following RTA, the leaking defect in the sphenoid sinus was associated with a pseudo-aneurysm of the internal carotid artery that was coiled before repair (Fig. 4). The mean duration of postoperative hospitalization was 6 days (range 4–8 days). None of the patients developed significant intra- or postoperative complications during
123
their hospital stay. Transient headaches and smell changes were reported in 18 and 12 cases respectively, and they all resolved after 3 months. The mean follow-up duration was 31.2 ± 11.4 months (range 16–48 months). Only one case, following RTA, needed revision surgery to repair a missed defect. A secondary frontal mucocele developed in one case following repair with middle turbinate flap (Fig. 5).
Discussion Our study describes our experience with endoscopic closure of 42 cases of post-traumatic CSF Leak. Several modalities of investigations were reported in literature to diagnose and localize the leak [10, 11]. In the present series, our diagnostic protocol included measurement of
Eur Arch Otorhinolaryngol Fig. 4 A 15-year-old male with post-traumatic CSF leak following RTA. a Coronal CT scan showing fractured posterolateral wall and roof of left sphenoid sinus. b Axial T2weighted MRI showing bright CSF signal in the left sphenoid and maxillary sinuses (black asterisk). (c and d) Axial and sagittal CT angioscans showing coiled internal carotid pseudoaneurysm (white arrowhead)
Fig. 5 Coronal (a) and Sagittal (b) CT scans showing frontal mucocele (asterisk) due to obstruction of the frontal recess with a middle turbinate flap (white arrowhead)
Beta-2 transferrin and/or Beta-trace protein in fluid samples collected from the nose, multidetector computed tomography (MDCT) and high-resolution T2-weighted MRI. CT angiography was also requested for patients with sphenoid sinus fractures to exclude the development of pseudo-aneurysm of the internal carotid artery. Multilayer repair techniques using different tissue materials were described in literature [12–15]. In our study we mainly used fascia lata, fat, bone grafts, middle turbinate flap, and nasoseptal flap. Fibrin glue was used in all cases. Using the described techniques we were able to achieve a success rate of 97 %, and only one case needed a second repair. Good, though variable, success rates have been also reported in the literature confirming the validity of endoscopic repair of the
leaking defects [16–19]. A critical point in reconstruction is the elevation of mucosa around the edges of the defect for appropriate graft housing. In experimental studies, evidence has been found that a stable underlay graft becomes incorporated with the dura after 1 week [20]. Surgical interference for CSF leaks following RTA or other accidental trauma was done only for persistent leaks after failure of conservative treatment, as most of these leaks stop with conservative treatment. On the other hand, leaks following surgery on the paranasal sinuses, pituitary gland, or anterior skull base usually require immediate surgical repair since they are unlikely to spontaneously. Our series showed that this difference was most probably related to the larger size of the defects in iatrogenic cases.
123
Eur Arch Otorhinolaryngol
The requirement of postoperative lumbar drainage is controversial according to most series. The decision to use lumbar drainage is usually based upon several factors including the existing intracranial pathology and the size of the defect [12]. Lumbar drainage was needed in 4 cases in our series. Long-term follow-up, clinical and radiological, of these patients is very important. MRI is preferable to CT since the absence of complete re-ossification of the defect is not necessarily predictive of surgical failure. In our series we followed up the patients with MRI after 3 months and 1 year. Many important complications have been listed as a sequel of endonasal endoscopic CSF closure [13, 16]. In our series, none of the patients developed serious intra- or postoperative complications. Failures of skull base repair after endoscopic surgery can possibly be due to inappropriate localization of the defect, inadequate positioning of the grafting materials, tension or displacement of the repair, the presence of a large or multiple defects, insufficient graft adherence and postoperative infection of the surgical field [21, 22]. An important issue is to prepare carefully the housing graft area by removing all the mucosa surrounding the defect to facilitate adhesions between the graft and the adjacent bone. Missing of a second defect is another important issue in post-traumatic cases, and it occurred only once in our series. Therefore, careful examination of the skull base for multiple defects is highly recommended in these post-traumatic cases. Care must be taken in placement of grafting material, especially nasoseptal flaps, as not to obstruct the frontal recess to avoid secondary mucocele formation. A previously reported study by Zweig et al. [23]. showed that the use of fixators did not influence the surgical outcome. However, we feel safer using tissue fixators, e.g., fibrin glue, because they help keeping the graft in place especially during the early postoperative period. In conclusion, Post-traumatic CSF leaks can be managed successfully via the endonasal endoscopic route using a multilayer repair technique, thus limiting the complications associated with external approaches. We believe that our results support the efficacy and safety of this approach, and we recommend it in most cases of post-traumatic cerebrospinal fluid leaks.
References 1. Hegazy HM, Carrau RL, Snyderman CH, Kassam A, Zweig J (2000) Transnasal endoscopic repair of cerebrospinal fluid rhinorrhea: a meta-analysis. Laryngoscope 110:1166–1172 2. Kerman M, Cirak B, Dagtekin A (2002) Management of skull base fractures. Neurosurg Q 12:23–41
123
3. Banks CA, Palmer JN, Chiu AG et al (2009) Endoscopic closure of CSF rhinorrhea: 193 cases over 21 years. Otolaryngol Head Neck Surg 140:826–833 4. Martin TJ, Loehrtl TA (2007) Endoscopic CSF leak repair. Curr Opin Otolaryngol Head Neck Surg 15:35–39 5. Prosser JD, Vender JR, Solares CA (2011) Traumatic cerebrospinal fluid leaks. Otolaryngol Clin N Am 44:857–873 6. Komatsu M, Komatsu F, Cavallo LM, Solari D, Stango V, Inoue T, Cappabianca P (2011) Purely endoscopic repair of traumatic cerebrospinal fluid rhinorrhea from the anterior skull base—case report. Neurol Med Chir (Tokyo) 51:222–225 7. Filho BCA, Butugan O, Pa´dua FGM, Voegels RL (2005) Endoscopic repair of CSF rhinorrhea: experience of 44 cases. Brazilian Journal of OtoRhinoLaryngiology 71:472–476 8. Gregor BH (2008) Values of beta-2 transferrin and beta-trace protein as markers for cerebrospinal fluid fistula. Rhinology 46(2):82–85 9. Placantonakis DG, Tabaee A, Anand VK, Hiltzik D, Schwartz TH (2007) Safety of low-dose intrathecal fluorescein in endoscopic cranial base surgery. Neurosurgery 61:161–165 10. Martin TJ, Loehrtl TA (2007) Endoscopic CSF leak repair. Curr Opin Otolaryngol Head Neck Surg 15:35–39 11. Locatelli D, Rampa F, Acchiardi I, Bignami M, De Bernardi F, Castelnuvo P (2006) Endoscopic endonasal approaches for repair of cerebrospinal fluid leaks: 9 year experience. Neurosurgery 58:246–257 12. Schmerber S, Righini C, Lavielle JP et al (2001) Endonasal endoscopic closure of cerebrospinal fluid rhinorrhea. Skull Base 11:47–58 13. Wormald PJ, McDonogh M (1997) ‘‘Bath-plug’’ technique for the endoscopic management of cerebrospinal fluid leaks. J Laryngol Otol 111:1042–1046 14. Chatrath P, Saleh HA (2006) Endoscopic repair of cerebrospinal fluid rhinorrhea using bone pate. Laryngoscope 116:1050–1054 15. Simmen D, Jones N (2005) Skull base surgery: management of skull base lesions with a CSF leak. In: Simmen D, Jones D (eds) Manual of endoscopic sinus surgery. Thieme, Stuttgart, pp 240–251 16. Senior BA, Jafry K, Benninger M (2001) Safety and efficacy of endoscopic repair of CSF leaks and encephaloceles: a survey of the members of the American Rhinologic Society. Am J Rhinol 15:21–25 17. Gilat H, Rappaport Z, Yaniv E (2011) Endoscopic transnasal Cerebrospinal Fluid leak repair: a 10 year experience. IMAJ 13:597–599 18. Martin TJ, Loehrtl TA (2007) Endoscopic CSF leak repair. Curr Opin Otolaryngol Head Neck Surg 15:35–39 19. Locatelli D, Rampa F, Acchiardi I, Bignami M, De Bernardi F, Castelnuvo P (2006) Endoscopic endonasal approaches for repair of cerebrospinal fluid leaks: 9 year experience. Neurosurgery 58:246–257 20. Hosemann W, Goede U, Sauer M (1999) Wound healing of mucosal autografts for frontal cerebrospinal fluid leaksVclinical and experimental investigations. Rhinology 37:108–112 21. Draf W, Schick B (2007) How I do it: endoscopic-microscopic anterior skull base reconstruction. Skull Base 17:53–58 22. Lund VJ (2002) Endoscopic management of cerebrospinal fluid leaks. Am J Rhinol 16:17–23 23. Zweig JL, Carrau C, Synderman R, Carrau P, Gardner A, Mintz A (2000) Endoscopic repair of cerebrospinal fluid leaks to the sinonasal tract: predictors of success. Otolaryngo Head Neck Surg. 123:195–201
RHINOLOGY
Endoscopic endonasal multilayer repair of traumatic CSF rhinorrhea Ahmed Aly Ibrahim1 • Mohamed Okasha2 • Samy Elwany1
Received: 18 December 2014 / Accepted: 31 May 2015 Ó Springer-Verlag Berlin Heidelberg 2015
Abstract The incidence of traumatic CSF has increased in recent years due to increased incidence of road traffic accidents (RTA) as well the increasing number of endoscopic sinus surgeries (ESS). The objective of this study is to present our experience in management of traumatic CSF leaks using the endoscopic multilayer repair technique. Forty-two patients (aged 10–75 years, 30 males and 12 females) presenting with confirmed post-traumatic CSF rhinorrhea were operated upon between January 2007 and December 2013. The endoscopic multilayer technique was used in all cases. Electromagnetic navigation was used in some cases. All cases presented with intermittent watery rhinorrhea. The duration of the rhinorrhea ranged from 3 days to 1 year before repair. One case presented after 10 years from the causative trauma. Ten cases had a history of meningitis. Nine cases had more than one defect. Iatrogenic defects were larger than defects following accidental trauma. Two cases, following RTA, developed pseudoaneurysm of internal carotid artery. Ten cases had associated pneumocephalus. The mean duration of postoperative hospitalization was 6 days (range 4–8 days). The mean follow-up duration was 31.2 ?/- 11.4 months (range 16–48 months). None of our patient developed serious intra- or postoperative complications. Only one case required another surgery to repair a missed second defect. Post-traumatic CSF leaks can be successfully managed via the endonasal endoscopic route using the multilayer repair & Samy Elwany [email protected] 1
Department of Otolaryngology, Alexandria Medical School, Alexandria, Egypt
2
Department of Neurosurgery, Damanhur Medical National Institute, Damanhur, Egypt
technique. It is important to look for multiple defects in these cases. CT angiography is recommended for traumatic leaks involving the lateral wall of the sphenoid sinus to diagnose or exclude the development of pseudo-aneurysm of the internal carotid artery. Keywords CSF leak ESS RTA Skull base Endoscopic Repair
Introduction CSF rhinorrhea is an abnormal release of cerebrospinal fluid (CSF) from the subarachnoid space into the nasal cavity or paranasal sinuses. Approximately 80 % of posttraumatic CSF leaks result from nonsurgical (accidental) trauma. More than 50 % of the leaks are evident within the first 2 days, 70 % within the first week, and almost all present within the first 3 months [1]. The site of CSF leaks depends upon the type of trauma. The most common sites of CSF rhinorrhea following accidental trauma are the sphenoid sinus (30 %), frontal sinus (30 %), and ethmoid/cribriform area (23 %) [2]. On the other hand, The most common sites of CSF leak following endoscopic sinus surgery (ESS) are the roof of anterior ethmoid especially at the level of the cribriform plate (80 %), followed the frontal sinus (8 %) and sphenoid sinus (4 %). CSF leaks following neurosurgical procedures are usually located in the sphenoid sinus (67 %) because of the high number of pituitary tumors that are addressed via transsphenoidal approach [3]. The most common presenting symptom of CSF leaks is intermittent watery rhinorrhea that is classically positional in nature. In clinically suspicious cases, laboratory analysis of CSF markers will confirm the diagnosis [4]. Most cases
123
Eur Arch Otorhinolaryngol
of CSF Rhinorrhea following accidental trauma stop with conservative treatment within few weeks after trauma. Iatrogenic cases are much less likely to heal spontaneously. The surgical approach to CSF leak repair has undergone a major progression. Multiple technical advancements have contributed to this progression, but the most important amongst all has been the development and improvement of the endoscopic technique for identification, localization and repair of the leaking defects. The endoscopic endonasal approach has been reported to be well tolerated and much safer than external approaches. Moreover, it has several other advantages in comparison to external techniques. These advantages include accurate localization of the defect, preservation of olfaction in most cases, no brain retraction especially on the frontal lobe, less morbidity and less hospital stay. However, several authors have reported variable success rates [5–7] apparently due to differences in their diagnostic protocols, repair techniques, and other factors. In this paper we present our experience in endoscopic repair of 42 cases of post-traumatic CSF rhinorrhea. All repairs were performed using the endoscopic endonasal multilayer technique.
Materials and methods The study was conducted on 42 cases of post-traumatic CSF rhinorrhea who were referred to the Department of Otolaryngology—Head and Neck Surgery between January 2007 and December 2013. An informed consent from all patients was taken after they were fully informed about the surgical procedure and the use of their data in the research.
Preoperative evaluation and identification of the defect Careful history and complete general, otorhinolaryngological, and neurological examination were performed in all cases. Levels of beta-2 transferrin and/or beta-trace protein were measured in fluid samples collected from the nose in all cases. A cut-off point of 1.11 mg/ml for Betatrace protein was used to confirm the presence of CSF in the collected fluid [8]. Our routine radiologic protocol included multidetector computed tomography (MDCT) with ultra-thin 1 mm cuts (bone and soft tissue settings) and multiplanar reconstruction, and magnetic resonance (MRI) cisternography using high-resolution T2-weighted sequence. CT angiography was requested for fractures involving the lateral wall of sphenoid sinus to exclude the development of pseudoaneurysm of the internal carotid artery.
123
Operative technique We used Intrathecal fluorescein in all patients to further confirm CSF leak, localize the defect, and check the success of the repair at the end of surgery [9]. Intraoperative navigation was used in some patients (GE InstaTrackTM 3500 electromagnetic navigation system). 0°, 30° and 45° endoscopes were used during surgery. The procedure started with identification of the site of leak. Partial middle turbinectomy was done, in most cases, to create a wide surgical corridor and for use of its bone as a graft in multilayer repair. The mucoperiosteum around the defect was removed to prepare the area for grafting. A multilayer (sandwich grafting) reconstruction was performed using some or all of the following 5 layers according to the size, site, and etiology of the defect. 1st layer; fat plug harvested from the anterior aspect of the thigh (intracranial/intradural) 2nd layer; fascia lata tucked underneath the bony edge (intracranial/extradural). 3rd layer; bone harvested from the middle turbinate or nasal septum (intracranial/extradural). 4th layer; overlay graft over the defect using fascia lata (extracranial). 5th layer; free mucoperiosteal graft or vascularized nasoseptal flap or middle turbinate flap (extracranial). The underlay fascia lata graft was fashioned so that it had at least 1 mm of diameter extending circumferentially beneath the edges of the bony defect. Underlay grafting was not performed for defects in the cribriform plate. Other defects larger than 5 mm in diameter usually required the 5 layers of repair. Table 1 summarizes the tissues used for the repair. The grafts were stabilized in position with fibrin glue. Lack of fluorescein flow from the repaired defects was used to confirm achievement of watertight ‘gasket-seal’ at the end of the operation. The reconstruction area was further secured with SurgifloÒ, SurgicelÒ, and/or GelfoamÒ. The nose was then packed with MerocelÒ and the pack is left for 3–4 days postoperatively. In cases of sphenoid sinus defect the steps were modified, where exploration of anterior skull base was not mandatory. An angled 45° endoscope is used to visualize the lateral recess and confirm the presence of a CSF leak. In cases of post-hypophysectomy CSF leak following transsphenoidal surgery, repair was done aiming to close Table 1 Tissues used for CSF leakage repair (42 patients) Tissue used
No of patients (%)
Fascia lata
42 (100 %)
Fat plug
35 (83.3 %)
Bone graft
31 (73.8 %)
Nasoseptal flap
29 (69 %)
Middle turbinate flap
3 (7.14 %)
Eur Arch Otorhinolaryngol
the diaphragmatic arachnoid defect with absorbable material and to pack the sella and sphenoid sinus with fat patch rapped in SurgicelÒ supporting the repair in place. After packing of the sella, a layer of fascia lata was applied intrasellar in underlay fashion bellow the edges of the dura and another layer onlay covering the sella floor over the dural edges in the roof of the sphenoid sinus, then packing of the sphenoid sinus with fat. Nasoseptal flap was in all cases to cover the sellar defect. SurgicelÒ was applied to the edges of the nasoseptal flap then a layer of fibrin glue covered the flap. No intraoperative complications were encountered. Postoperative care included antibiotic prophylaxis, elevation of head of bed, food softeners, nasal packing (3–7 days) in all cases, and lumbar drainage in four cases. Hospital postoperative stay ranged from 2 to 7 days. The mean follow-up duration was 31.2 ± 11.4 months (range 16–48 months). Radiological follow-up (MRI) was done after 3 months and 1 year. The repair was considered
successful if there was no CSF leak at the end of the follow-up period.
Results The current study was carried out on 42 cases of traumatic CSF rhinorrhea. Thirty patients were males (71.4 %) and 12 females (28.6 %). Age varied from 10 to 75 years. Beta2 transferrin and/or beta-trace protein measurements confirmed the presence of CSF in the sample fluid of all patients. All cases presented with intermittent watery rhinorrhea. The duration of the rhinorrhea ranged from 3 days to 1 year before repair. One case presented after 10 years from the causative trauma (Fig. 1). 12 cases presented with headache, 8 cases with hyposmia/anosmia, and 10 cases with history of meningitis. Table 2 summarizes the etiology and some of the characteristics of the leaking defects. CSF leakage occurred
Fig. 1 A 36-year-old male with late post-traumatic CSF leak (10 years after trauma). (a and b) Coronal CT scans showing the defect in the posterior wall of right frontal sinus. (c and d) Coronal T2-weighted MRI showing frontal lobe encephalomalacia and leaking CSF into the right nasal cavity
Table 2 Etiology of CSF leakage and some characteristics of the leaking defect Etiology
Number of patients
Number of defects [1
1
Size of defect
Total
Mean site of defect (mm) CP/ASB
Pneumocephalus on CT
Sphenoid sinus
Accidental trauma
10
4
6
16
5.9
12
4
2
ESS
12
10
2
14
7.2
14
0
1
Hypophysectomy
15
15
0
15
12.3
0
15
4
5
4
1
6
9.5
6
0
3
Anterior skull base surgery
CP cribriform plate, ASB anterior skull base
123
Eur Arch Otorhinolaryngol Fig. 2 CSF leakage following endoscopic sinus surgery. a Coronal CT scan showing a big defect in the posterior cribriform plate and ethmoid roof (white arrowhead) with underlying fluid attenuation and in the right ethmoid and maxillary sinuses. b Coronal MRI showing bright CSF signal at the site of the leak (white arrowhead) as well as in the right ethmoid and maxillary sinuses
Fig. 3 Post-Hypophysectomy CSF leak. Sagittal a and Coronal b T2-weighted MRI showing bright CSF signal in the sella and sphenoid sinus (black asterisk). A frontal pneumocephalus is also seen (white asterisk) in the sagittal section
after RTA (7 cases), falling from height (3 cases), endoscopic sinus surgery (12 cases), hypophysectomy (15 cases), and skull base procedures (5 cases). More than one defect was seen in six out of 10 cases (60 %) following accidental trauma, two out of 12 cases (16.6 %) following endoscopic sinus surgery, and one out of five cases following anterior skull base surgery. Post-surgical defects (Fig. 2) tended to be larger than defects caused by accidental trauma. Pneumocephalus was detected in 10 cases (Fig. 3). In two cases following RTA, the leaking defect in the sphenoid sinus was associated with a pseudo-aneurysm of the internal carotid artery that was coiled before repair (Fig. 4). The mean duration of postoperative hospitalization was 6 days (range 4–8 days). None of the patients developed significant intra- or postoperative complications during
123
their hospital stay. Transient headaches and smell changes were reported in 18 and 12 cases respectively, and they all resolved after 3 months. The mean follow-up duration was 31.2 ± 11.4 months (range 16–48 months). Only one case, following RTA, needed revision surgery to repair a missed defect. A secondary frontal mucocele developed in one case following repair with middle turbinate flap (Fig. 5).
Discussion Our study describes our experience with endoscopic closure of 42 cases of post-traumatic CSF Leak. Several modalities of investigations were reported in literature to diagnose and localize the leak [10, 11]. In the present series, our diagnostic protocol included measurement of
Eur Arch Otorhinolaryngol Fig. 4 A 15-year-old male with post-traumatic CSF leak following RTA. a Coronal CT scan showing fractured posterolateral wall and roof of left sphenoid sinus. b Axial T2weighted MRI showing bright CSF signal in the left sphenoid and maxillary sinuses (black asterisk). (c and d) Axial and sagittal CT angioscans showing coiled internal carotid pseudoaneurysm (white arrowhead)
Fig. 5 Coronal (a) and Sagittal (b) CT scans showing frontal mucocele (asterisk) due to obstruction of the frontal recess with a middle turbinate flap (white arrowhead)
Beta-2 transferrin and/or Beta-trace protein in fluid samples collected from the nose, multidetector computed tomography (MDCT) and high-resolution T2-weighted MRI. CT angiography was also requested for patients with sphenoid sinus fractures to exclude the development of pseudo-aneurysm of the internal carotid artery. Multilayer repair techniques using different tissue materials were described in literature [12–15]. In our study we mainly used fascia lata, fat, bone grafts, middle turbinate flap, and nasoseptal flap. Fibrin glue was used in all cases. Using the described techniques we were able to achieve a success rate of 97 %, and only one case needed a second repair. Good, though variable, success rates have been also reported in the literature confirming the validity of endoscopic repair of the
leaking defects [16–19]. A critical point in reconstruction is the elevation of mucosa around the edges of the defect for appropriate graft housing. In experimental studies, evidence has been found that a stable underlay graft becomes incorporated with the dura after 1 week [20]. Surgical interference for CSF leaks following RTA or other accidental trauma was done only for persistent leaks after failure of conservative treatment, as most of these leaks stop with conservative treatment. On the other hand, leaks following surgery on the paranasal sinuses, pituitary gland, or anterior skull base usually require immediate surgical repair since they are unlikely to spontaneously. Our series showed that this difference was most probably related to the larger size of the defects in iatrogenic cases.
123
Eur Arch Otorhinolaryngol
The requirement of postoperative lumbar drainage is controversial according to most series. The decision to use lumbar drainage is usually based upon several factors including the existing intracranial pathology and the size of the defect [12]. Lumbar drainage was needed in 4 cases in our series. Long-term follow-up, clinical and radiological, of these patients is very important. MRI is preferable to CT since the absence of complete re-ossification of the defect is not necessarily predictive of surgical failure. In our series we followed up the patients with MRI after 3 months and 1 year. Many important complications have been listed as a sequel of endonasal endoscopic CSF closure [13, 16]. In our series, none of the patients developed serious intra- or postoperative complications. Failures of skull base repair after endoscopic surgery can possibly be due to inappropriate localization of the defect, inadequate positioning of the grafting materials, tension or displacement of the repair, the presence of a large or multiple defects, insufficient graft adherence and postoperative infection of the surgical field [21, 22]. An important issue is to prepare carefully the housing graft area by removing all the mucosa surrounding the defect to facilitate adhesions between the graft and the adjacent bone. Missing of a second defect is another important issue in post-traumatic cases, and it occurred only once in our series. Therefore, careful examination of the skull base for multiple defects is highly recommended in these post-traumatic cases. Care must be taken in placement of grafting material, especially nasoseptal flaps, as not to obstruct the frontal recess to avoid secondary mucocele formation. A previously reported study by Zweig et al. [23]. showed that the use of fixators did not influence the surgical outcome. However, we feel safer using tissue fixators, e.g., fibrin glue, because they help keeping the graft in place especially during the early postoperative period. In conclusion, Post-traumatic CSF leaks can be managed successfully via the endonasal endoscopic route using a multilayer repair technique, thus limiting the complications associated with external approaches. We believe that our results support the efficacy and safety of this approach, and we recommend it in most cases of post-traumatic cerebrospinal fluid leaks.
References 1. Hegazy HM, Carrau RL, Snyderman CH, Kassam A, Zweig J (2000) Transnasal endoscopic repair of cerebrospinal fluid rhinorrhea: a meta-analysis. Laryngoscope 110:1166–1172 2. Kerman M, Cirak B, Dagtekin A (2002) Management of skull base fractures. Neurosurg Q 12:23–41
123
3. Banks CA, Palmer JN, Chiu AG et al (2009) Endoscopic closure of CSF rhinorrhea: 193 cases over 21 years. Otolaryngol Head Neck Surg 140:826–833 4. Martin TJ, Loehrtl TA (2007) Endoscopic CSF leak repair. Curr Opin Otolaryngol Head Neck Surg 15:35–39 5. Prosser JD, Vender JR, Solares CA (2011) Traumatic cerebrospinal fluid leaks. Otolaryngol Clin N Am 44:857–873 6. Komatsu M, Komatsu F, Cavallo LM, Solari D, Stango V, Inoue T, Cappabianca P (2011) Purely endoscopic repair of traumatic cerebrospinal fluid rhinorrhea from the anterior skull base—case report. Neurol Med Chir (Tokyo) 51:222–225 7. Filho BCA, Butugan O, Pa´dua FGM, Voegels RL (2005) Endoscopic repair of CSF rhinorrhea: experience of 44 cases. Brazilian Journal of OtoRhinoLaryngiology 71:472–476 8. Gregor BH (2008) Values of beta-2 transferrin and beta-trace protein as markers for cerebrospinal fluid fistula. Rhinology 46(2):82–85 9. Placantonakis DG, Tabaee A, Anand VK, Hiltzik D, Schwartz TH (2007) Safety of low-dose intrathecal fluorescein in endoscopic cranial base surgery. Neurosurgery 61:161–165 10. Martin TJ, Loehrtl TA (2007) Endoscopic CSF leak repair. Curr Opin Otolaryngol Head Neck Surg 15:35–39 11. Locatelli D, Rampa F, Acchiardi I, Bignami M, De Bernardi F, Castelnuvo P (2006) Endoscopic endonasal approaches for repair of cerebrospinal fluid leaks: 9 year experience. Neurosurgery 58:246–257 12. Schmerber S, Righini C, Lavielle JP et al (2001) Endonasal endoscopic closure of cerebrospinal fluid rhinorrhea. Skull Base 11:47–58 13. Wormald PJ, McDonogh M (1997) ‘‘Bath-plug’’ technique for the endoscopic management of cerebrospinal fluid leaks. J Laryngol Otol 111:1042–1046 14. Chatrath P, Saleh HA (2006) Endoscopic repair of cerebrospinal fluid rhinorrhea using bone pate. Laryngoscope 116:1050–1054 15. Simmen D, Jones N (2005) Skull base surgery: management of skull base lesions with a CSF leak. In: Simmen D, Jones D (eds) Manual of endoscopic sinus surgery. Thieme, Stuttgart, pp 240–251 16. Senior BA, Jafry K, Benninger M (2001) Safety and efficacy of endoscopic repair of CSF leaks and encephaloceles: a survey of the members of the American Rhinologic Society. Am J Rhinol 15:21–25 17. Gilat H, Rappaport Z, Yaniv E (2011) Endoscopic transnasal Cerebrospinal Fluid leak repair: a 10 year experience. IMAJ 13:597–599 18. Martin TJ, Loehrtl TA (2007) Endoscopic CSF leak repair. Curr Opin Otolaryngol Head Neck Surg 15:35–39 19. Locatelli D, Rampa F, Acchiardi I, Bignami M, De Bernardi F, Castelnuvo P (2006) Endoscopic endonasal approaches for repair of cerebrospinal fluid leaks: 9 year experience. Neurosurgery 58:246–257 20. Hosemann W, Goede U, Sauer M (1999) Wound healing of mucosal autografts for frontal cerebrospinal fluid leaksVclinical and experimental investigations. Rhinology 37:108–112 21. Draf W, Schick B (2007) How I do it: endoscopic-microscopic anterior skull base reconstruction. Skull Base 17:53–58 22. Lund VJ (2002) Endoscopic management of cerebrospinal fluid leaks. Am J Rhinol 16:17–23 23. Zweig JL, Carrau C, Synderman R, Carrau P, Gardner A, Mintz A (2000) Endoscopic repair of cerebrospinal fluid leaks to the sinonasal tract: predictors of success. Otolaryngo Head Neck Surg. 123:195–201
