- Email: [email protected]
Determining leak locations during transnasal endoscopic repair of cerebrospinal fluid rhinorrhea
Auris Nasus Larynx 38 (2011) 335–339 www.elsevier.com/locate/anl
Determining leak locations during transnasal endoscopic repair of cerebrospinal fluid rhinorrhea Li Qiao 1, Tao Xue 1,*, Ding-jun Zha 1, Fu-quan Chen, Xu Li, Jian-hua Qiu *, Zhao-hui Shi, Li-ting Wen Department of Otolaryngology-Head and Neck Surgery, Xijing Hospital, Fourth Military Medical University, No. 15, Changle Western Road, Xi’an 710032, China Received 6 May 2009; accepted 30 November 2010 Available online 12 January 2011
Abstract Objective: For transnasal endoscopic repair procedures to be successful, it is critical to identify leak locations during surgery. We aim to evaluate different methods to more accurately detect leak locations during the endoscopic repair of cerebrospinal fluid rhinorrhea. Materials and methods: We performed a retrospective chart review of 39 cases undergoing endoscopic repair of cerebrospinal fluid rhinorrhea. The leak locations were determined using preoperative nasal endoscopy, radioisotope scanning, the intraoperative image-guided system, and intraspinal normal saline injection. Results: The cerebrospinal fluid leak location was in the sphenoidal sinus in 9 cases, the ethmoid sinus in 17 cases, and in the frontal sinus in 1 case. The leak locations could not be determined in the remaining 12 cases using this method alone. For these 12 cases, after the ethmoid sinus was opened and the lateral wall of sphenoidal sinus was exposed with the aid of the intraoperative image-guided system, outflow of cerebrospinal fluid was present on the lateral wall of sphenoidal sinus (in 1 case) and on the ethmoid roof (in 3 cases). Furthermore, using intraspinal saline injection (20–30 ml), leak locations were detected in the sphenoidal sinus (2 cases) and in ethmoid sinus (6 cases) of the remaining cases. Conclusion: For cerebrospinal fluid rhinorrhea patients whose leak locations are difficult to determine, surgeons can increase their operative success rates by performing radioisotope scanning and intraspinal saline injections and by using image-guided surgical systems. These safe and effective methods can be used to successfully detect leak locations during transnasal endoscopic repair of cerebrospinal fluid leaks. # 2011 Published by Elsevier Ireland Ltd. Key words: Endoscopy; Cerebrospinal fluid rhinorrhea; Skull base
1. Introduction Cerebrospinal fluid rhinorrhea is typically traumatic or spontaneous. Traumatic cerebrospinal fluid rhinorrhea usually occurs in the iatrogenic anterior skull base [1]. The etiology of spontaneous cerebrospinal fluid rhinorrhea is uncertain, but it is generally associated with developmental defects of the skull base and intracranial hyperten-
* Corresponding author. Tel.: +86 29 84775381; fax: +86 29 84773427. E-mail addresses: [email protected] (T. Xue), [email protected] (J.-h. Qiu). 1 These author contributed equally to this study, should be regarded as joint first author. 0385-8146/$ – see front matter # 2011 Published by Elsevier Ireland Ltd. doi:10.1016/j.anl.2010.11.001
sion [2,3]. Because of the risk of complications such as meningitis, brain abscess, and pneumocephalus, all persistent CSF leaks should be repaired. The traditional surgical procedures to treat cerebrospinal fluid rhinorrhea such as frontal craniotomy are not only dangerous, but they may also lead to complications, such as reduced osmesthesia, intracranial hemorrhage, and pneumatosis. Frontal craniotomy has been the traditional surgical approach. The recent involvement of transnasal endoscopic sinus surgery and microsurgery has made it possible to treat cerebrospinal fluid oto- and rhinorrhea with a high degree of success and less operative morbidity. At present, the best way to repair cerebrospinal fluid rhinorrhea is transnasal endoscopic repair, because it possesses safe, reliable, minimally invasive
336
L. Qiao et al. / Auris Nasus Larynx 38 (2011) 335–339
and highly successful advantages and has been widely used all over the world. Transnasal endoscopic repair of cerebrospinal fluid rhinorrhea has been accepted by many otolaryngologists because it is minimally invasive, safe, quick, and accurate [4–7]. The repair methods include overlay and underlay techniques, and ‘‘button’’ and ‘‘bath-plug’’ closures [8,9]. The most commonly used repair materials are nasal mucosa, periosteum, fascia lata, muscle, fat, and artificial materials [10–13]. The success rates of transnasal endoscopic repair range between 76% and 97% [14]. Cerebrospinal fluid rhinorrhea can be definitively diagnosed by nasal endoscopy, determining the glucose content, biochemically measuring B2-transferrin, and by using computed tomography (CT) and magnetic resonance imagining (MRI) in most patients. The leak locations may be estimated by the cerebrospinal fluid outflow during surgery. In some patients, however, it is difficult to determine the leak locations because there may be less rhinorrhea, inflammatory swelling around the leak, or obstruction of the granulation tissue. It is a critical step of successful surgery to find leak locations during operations. For transnasal endoscopic repair procedure to be successful, it is critical to identify the leak locations during surgery. In this study, we used a combination of methods to detect the leak locations, including preoperative nasal endoscopy, radioisotope scanning, the intraoperative navigation system, and intraspinal saline injection. Our aim was to more accurately determine the leak locations to improve operative success rates.
2. Materials and methods 2.1. Subjects All study methods were approved by the Clinical Investigation Committee of the Xi Jing Hospital, Fourth Military Medical University. Between August 2001 and January 2006, 39 patients, were diagnosed with cerebrospinal fluid rhinorrhea by preoperative nasal endoscopy, radioisotope scanning, and glucose content determinations. The patients underwent transnasal endoscopic repair of cerebrospinal fluid rhinorrhea in our hospital. The patients ranged in age from 9 to 66 years (mean age of 36 years). Twenty patients were men and 19 were women. The locations of the leaks were determined prior to the operation in 27 cases. For the remaining 12 cases, we used a navigation system and intraspinal saline injection to find the leak locations during surgery. The patients’ general characteristics are shown in Table 1.
Table 1 All patient’s general condition. Items
Results (n = 39)
Age
Mean: 36 years Range: 9–66 years
Gender
Male: 20 Female: 19
Etiology of leak
Sinus surgery: 3 Trauma: 8 Spontaneous: 26 Skull base surgery 2
Duration of leak before surgery
Mean: 3.2 years Median: 2.5 years Range: 0–11 years
be determined by nasal endoscopy, preoperative radioisotope scanning was performed by injecting 99mTc-DTPA (diethylenetriaminepentaacetic acid) into the spinal subarachnoid space. Once the 99mTc-DTPA entered the cerebrospinal fluid pathway, 99mTc-DTPA scintigraphy was performed to determine whether cerebrospinal fluid rhinorrhea was present. 2.2.2. Navigation system One or two days prior to the operation, axial spiral CT was used to scan the nasal sinuses. The CT data were stored in the image-guided surgical system of the German Brain Lab for further use. Following anesthesia, patients were fitted with a head frame fixed with an infrared reflection sphere. The Z-Touch laser registration method was used, which had a mean error of 0.8 mm (0.5–1.2 mm). In all cases, the contouring of the ethmoid sinus was performed under image guidance. Cells in the ethmoid sinus were removed to expose the entire ethmoid roof. Alternatively, the lateral wall of the sphenoidal sinus was exposed by opening the sphenoidal sinus from its natural ostium. For the patients whose leak locations was undetected, intraspinal saline injection was then used to find the leak locations.
2.2. Methods
2.2.3. Intraspinal normal saline injection All patients lay on their sides and flexing their knee. A lumbar puncture was performed, and a spinal catheter that was connected to a pressure transducer (SONY Co.) was fixed. The cerebrospinal fluid pressure was determined before the operation. For the patients with leak locations that could not be determined by opening the sphenoidal or ethmoid sinuses, normal saline (in 20 ml volumes) was injected into the vertebral canal to observe the outflow of cerebrospinal fluid. When each 20-ml volume of normal saline was injected, the alteration of cerebrospinal fluid pressure was assessed using the pressure transducer to prevent a rapid increase in cerebrospinal fluid pressure.
2.2.1. Radioisotope scanning All cases (n = 39) first underwent nasal endoscopy to determine the leak locations. If the leak locations could not
2.2.4. Repairing leaks Small defects (<5 mm) were repaired with autologous muscle from the thigh using nasal endoscopy, but larger
L. Qiao et al. / Auris Nasus Larynx 38 (2011) 335–339
skull-base defects or skull-base developmental malformations were repaired using the intracranial-nasal approach [18]. 2.2.5. Postsurgical care Patients were positioned in a semi-reclining position and were intravenously administered 25% mannite (125 ml) and furosemide (20 mg) twice daily for 7–10 days. The iodoform gauze was drawn out of the nasal cavity 7 days following surgery, and the patients were discharged 10 days after the operation.
3. Results Of the 39 cases, the leak locations were determined by nasal endoscopy in 27 cases, but could not be determined for the remaining 12 cases. Twelve patients whose leak locations could not be determined by preoperative nasal endoscopy and radioisotope scanning underwent endoscopic repair of the cerebrospinal fluid leaks with the assistance of the image-guided system. The leak locations were suspected to be in the sphenoidal sinus in 3 cases. In one of these cases, after removing the granulation tissue, a small fissure was present on the posterior and superior wall of the sphenoidal sinus, and the image-guided system suggested that the fissure communicated with intracranial space. In the other 2 cases, under image guidance, after the edema mucosa was gently teased up, and 20 ml of saline was injected into vertebral canal, there was outflow present on lateral wall of the sphenoidal sinus. In 9 cases, the leaks were suspected to be located in the ethmoid sinus. After contouring the ethmoid sinus, the leak locations were well exposed in 3 cases. In the other 6 cases, there was no outflow of cerebrospinal fluid after contouring the ethmoid sinus. When these patients lay in the head-down tilt position and intracranial pressure was increased by pressing the internal jugular vein, fluid was still not present. Lastly, normal saline was injected into the vertebral canal (20 ml in 3 cases, 28 ml in 1 case, and 35 ml in 2 cases). Outflow was observed from a small fissure in the middle of the ethmoid roof in 4 cases. In 1 patient with substantial granulation tissue in the meninges, the leak location was in the posterior ethmoid sinus and close to the sphenoidal sinus. In the last case, the fluid was leaking from the cribriform plate, and the image-guided system suggested that the patient had a bony fissure defect. Details are shown in Tables 2 and 3. Follow-ups were performed 3–56 months (average 46 months) after surgery. Of the 39 cases, the first operation was successful in 36 cases, and leakage recurred in 3 cases.
337
Table 2 Leak locations as detected by different methods before the operations. Methods
Items
Results
Endoscope (n = 39)
Cerebrospinal fluid rhinorrhea
Yes: 27
Side of leak
Location of leak
Isotope scanning (n = 12)
Cerebrospinal fluid rhinorrhea Side of leak
Location of leak
Preoperative radioisotope scanning is a simple, safe, and reliable diagnostic method for cerebrospinal fluid rhinor-
Yes: 12 Uncertain: 0 Left: 0 Right: 0 Both: 0 Unknown: 12 Fovea ethmoigails: 0 Cribriform plate: 0 Sphenoid: 0 Frontal recess: 0 Unknown: 12
rhea. In patients where cerebrospinal fluid rhinorrhea is highly suspected, preoperative radioisotope scanning may be used to make a definitive diagnosis. The disadvantage of using preoperative radioisotope scanning is that the technique cannot be used to determine leak locations. The image-guided surgical system creates three-dimensional computerized reconstructions from preoperative CT and MRI images. Using this system, the surgeon can view the precise locations of surgical appliances in the surgical Table 3 Leak locations as detected by different methods during the operations. Methods
Items
Results
Image guidance (n = 4)
Side of leak
Left: 1
Location of leak
Intraspinal injection saline (n = 8)
Side of leak
Location of leak
4. Discussion
Uncertain: 12 Left: 7 Right: 20 Both: 0 Unknown: 12 Fovea ethmoigails: 17 Cribriform plate: 0 Sphenoid: 9 Frontal recess: 1 Unknown: 12
Right: 3 Both: 0 Unknown: 0 Fovea ethmoigails: 2 Cribriform plate: 1 Sphenoid: 1 Frontal recess: 0 Unknown: 0 Left: 3 Right: 5 Both: 0 Unknown: 0 Fovea ethmoigails: 5 Cribriform plate: 1 Sphenoid: 2 Frontal recess: 0 Unknown: 0
338
L. Qiao et al. / Auris Nasus Larynx 38 (2011) 335–339
field using the computer’s locating system. This surgical system has been widely used to perform a variety of nasal sinus and anterior skull base surgeries [15,16]. The imageguided system has the following advantages for repairing cerebrospinal fluid leaks. (1) It can be used to precisely estimate anatomical structures (particularly useful for the anterior wall and lateral wall of the sphenoidal sinus), and avoid the internal carotid artery and optic nerve. (2) After complete contouring of the ethmoid sinus by thoroughly removing cells on the ethmoid roof, it can detect leak locations easily. (3) The system can be used to visualize bony defects that cannot be viewed in the preoperative CT examination, which makes surgeries more accurate. Repair of cerebrospinal fluid leaks in the sphenoidal sinus is difficult, especially when they are located on the lateral wall because the surgical vision field is poor and the lateral wall of the sphenoidal sinus is adjacent to important structures, including the internal carotid artery and optic nerve. We used the image-guided surgical system to accurately localize the anterior and lateral wall of the sphenoidal sinus. Additionally, the system allowed the optic nerve and internal carotid artery to be visualized so the surgeon could avoid these important structures. The visualization of bony defects in the three-dimensional reconstruction increased surgical safety and the accuracy of localizing leaks, which improved the surgeons’ confidence. The leaks were accurately localized using the above approaches in our 39 cases, which improved the success rate of the surgeries. Simple nasal endoscopic surgery is not suitable for larger skull-base defects or skull-base malformations. Of 39 patients, 2 cases underwent intracranial-nasal repair (both with spontaneous cerebrospinal fluid leaks and meningoencephaloceles, one case had a 0.8 cm 1.4 cm skull-base defect and the other had a skull-base conical malformation), and the first repair was successful. Three patients experienced recurrence of the cerebrospinal fluid leaks, and all three were treated by reoperation. In one of the recurrent cases, although the leak was repaired by nasal endoscopy, when the intracranial pressure was increased, the cerebrospinal fluid leak reoccurred because of multiple bony defects on the cribriform plate. A second intracranial repair was successfully performed to treat the patient. Therefore, it is critical to determine the leak locations and select the appropriate surgical procedures to improve the success rates for nasal endoscopic repair of cerebrospinal fluid leaks. To repair traumatic or spontaneous cerebrospinal fluid leaks, one of the greatest challenges is detecting the precise leak location. Generally, CT and MRI are not very useful for determining the leak location. Intraoperative outflow of cerebrospinal fluid is decreased and sometimes is not present due to the patients’ position when they are lying down, which makes it difficult to determine the leak location. When the leak locations cannot be detected, most surgeons inject fluorescein into the vertebral canal to observe the
outflow of fluorescein though the nasal endoscope [17]. This method is indeed practicable, but the high concentration of fluorescein may contribute to nerve injury. We injected saline into the vertebral canal to increase cerebrospinal fluid pressure, which resulted in a visible outflow of cerebrospinal fluid, which could be clearly observed under the nasal endoscope. During the endoscopic repair of the cerebrospinal fluid leak, we injected 15–35 ml of normal saline into the vertebral canal in 8 cases, and all of the leak locations were well exposed. Monitoring cerebrospinal fluid pressure during surgery is important because most patients with spontaneous cerebrospinal fluid leaks also have high cerebrospinal fluid pressure.
5. Conclusion For cerebrospinal fluid rhinorrhea patients whose leak locations are difficult to determine, surgeons can better determine the orientation of the surgical fields by performing radioisotope scanning and intraspinal saline injections and by using image-guided surgical systems. These safe and effective methods can be used to successfully detect leak locations during transnasal endoscopic repair of cerebrospinal fluid leaks.
References [1] Kelley TF, Stankiewicz JA, Chow JM, Origitano TC, Shea J. Endoscopic closure of postsurgical anterior cranial fossa cerebrospinal fluid leaks. Neurosurgery 1996;39:743–6. [2] Lopatin AS, Kapitanov DN, Potapov AA. Endonasal endoscopic repair of spontaneous cerebrospinal fluid leaks. Arch Otolaryngol Head Neck Surg 2003;129:859–63. [3] Carrau RL, Snyderman CH, Kassam AB. The management of cerebrospinal fluid leaks in patients at risk for high-pressure hydrocephalus. Laryngoscope 2005;115:205–12. [4] Briggs RJ, Wormald PJ. Endoscopic transnasal intradural repair of anterior skull base cerebrospinal fluid fistulae. J Clin Neurosci 2004;11:597–9. [5] Bernal-Sprekelsen M, Alobid I, Mullol J, Trobat F, Toma´s-Barbera´n M. Closure of cerebrospinal fluid leaks prevents ascending bacterial meningitis. Rhinology 2005;43:277–81. [6] McMains KC, Gross CW, Kountakis SE. Endoscopic management of cerebrospinal fluid rhinorrhea. Laryngoscope 2004;114:1833–7. [7] Lee TJ, Huang CC, Chuang CC, Huang SF. Transnasal endoscopic repair of cerebrospinal fluid rhinorrhea and skull base defect: ten-year experience. Laryngoscope 2004;114:1475–81. [8] El-Banhawy OA, Halaka AN, El-Hafiz Shehab El-Dien A, Ayad H. Subcranial transnasal repair of cerebrospinal fluid rhinorrhea with free autologous grafts by the combined overlay and underlay techniques. Minim Invasive Neurosurg 2004;47:197–202. [9] Wormald PJ, McDonogh M. The bath-plug closure of anterior skull base cerebrospinal fluid leaks. Am J Rhinol 2003;17:299–305. [10] Lorenz RR, Dean RL, Hurley DB, Chuang J, Citardi MJ. Endoscopic reconstruction of anterior and middle cranial fossa defects using acellular dermal allograft. Laryngoscope 2003;113:496–501. [11] Weber SM, Kim J, Delashaw JB, Wax MK. Radial forearm free tissue transfer in the management of persistent cerebrospinal fluid leaks. Laryngoscope 2005;115:968–72.
L. Qiao et al. / Auris Nasus Larynx 38 (2011) 335–339 [12] Costantino PD, Hiltzik DH, Sen C, Friedman CD, Kveton JF, Snyderman CF, et al. Sphenoethmoid cerebrospinal fluid leak repair with hydroxyapatite cement. Arch Otolaryngol Head Neck Surg 2001; 127:588–93. [13] Kelly DF, Oskouian RJ, Fineman I. Collagen sponge repair of small cerebrospinal fluid leaks obviates tissue grafts and cerebrospinal fluid diversion after pituitary surgery. Neurosurgery 2001;49: 885–90. [14] Ismail AS, Costantino PD, Sen C. Transnasal transsphenoidal endoscopic repair of CSF leakage using multilayer acellular dermis. Skull Base 2007;17:125–32.
339
[15] Wise SK, DelGaudio JM. Computer-aided surgery of the paranasal sinuses and skull base. Expert Rev Med Devices 2005;2:395–408. [16] Hemmerdinger SA, Jacobs JB, Lebowitz RA. Accuracy and cost analysis of image-guided sinus surgery. Otolaryngol Clin North Am 2005;38:453–60. [17] Stammberger H, Greistorfer K, Wolf G, et al. Surgical occlusion of cerebrospinal fistulas of the anterior skull base using intrathecal sodium fluorescein. Laryngorhinootologie 1997;76:595–607. [18] Woodworth BA, Schlosser RJ, Faust RA, Bolger WE. Evolutions in the management of congenital intranasal skull base defects. Arch Otolaryngol Head Neck Surg 2004;130:1283–8.
Determining leak locations during transnasal endoscopic repair of cerebrospinal fluid rhinorrhea Li Qiao 1, Tao Xue 1,*, Ding-jun Zha 1, Fu-quan Chen, Xu Li, Jian-hua Qiu *, Zhao-hui Shi, Li-ting Wen Department of Otolaryngology-Head and Neck Surgery, Xijing Hospital, Fourth Military Medical University, No. 15, Changle Western Road, Xi’an 710032, China Received 6 May 2009; accepted 30 November 2010 Available online 12 January 2011
Abstract Objective: For transnasal endoscopic repair procedures to be successful, it is critical to identify leak locations during surgery. We aim to evaluate different methods to more accurately detect leak locations during the endoscopic repair of cerebrospinal fluid rhinorrhea. Materials and methods: We performed a retrospective chart review of 39 cases undergoing endoscopic repair of cerebrospinal fluid rhinorrhea. The leak locations were determined using preoperative nasal endoscopy, radioisotope scanning, the intraoperative image-guided system, and intraspinal normal saline injection. Results: The cerebrospinal fluid leak location was in the sphenoidal sinus in 9 cases, the ethmoid sinus in 17 cases, and in the frontal sinus in 1 case. The leak locations could not be determined in the remaining 12 cases using this method alone. For these 12 cases, after the ethmoid sinus was opened and the lateral wall of sphenoidal sinus was exposed with the aid of the intraoperative image-guided system, outflow of cerebrospinal fluid was present on the lateral wall of sphenoidal sinus (in 1 case) and on the ethmoid roof (in 3 cases). Furthermore, using intraspinal saline injection (20–30 ml), leak locations were detected in the sphenoidal sinus (2 cases) and in ethmoid sinus (6 cases) of the remaining cases. Conclusion: For cerebrospinal fluid rhinorrhea patients whose leak locations are difficult to determine, surgeons can increase their operative success rates by performing radioisotope scanning and intraspinal saline injections and by using image-guided surgical systems. These safe and effective methods can be used to successfully detect leak locations during transnasal endoscopic repair of cerebrospinal fluid leaks. # 2011 Published by Elsevier Ireland Ltd. Key words: Endoscopy; Cerebrospinal fluid rhinorrhea; Skull base
1. Introduction Cerebrospinal fluid rhinorrhea is typically traumatic or spontaneous. Traumatic cerebrospinal fluid rhinorrhea usually occurs in the iatrogenic anterior skull base [1]. The etiology of spontaneous cerebrospinal fluid rhinorrhea is uncertain, but it is generally associated with developmental defects of the skull base and intracranial hyperten-
* Corresponding author. Tel.: +86 29 84775381; fax: +86 29 84773427. E-mail addresses: [email protected] (T. Xue), [email protected] (J.-h. Qiu). 1 These author contributed equally to this study, should be regarded as joint first author. 0385-8146/$ – see front matter # 2011 Published by Elsevier Ireland Ltd. doi:10.1016/j.anl.2010.11.001
sion [2,3]. Because of the risk of complications such as meningitis, brain abscess, and pneumocephalus, all persistent CSF leaks should be repaired. The traditional surgical procedures to treat cerebrospinal fluid rhinorrhea such as frontal craniotomy are not only dangerous, but they may also lead to complications, such as reduced osmesthesia, intracranial hemorrhage, and pneumatosis. Frontal craniotomy has been the traditional surgical approach. The recent involvement of transnasal endoscopic sinus surgery and microsurgery has made it possible to treat cerebrospinal fluid oto- and rhinorrhea with a high degree of success and less operative morbidity. At present, the best way to repair cerebrospinal fluid rhinorrhea is transnasal endoscopic repair, because it possesses safe, reliable, minimally invasive
336
L. Qiao et al. / Auris Nasus Larynx 38 (2011) 335–339
and highly successful advantages and has been widely used all over the world. Transnasal endoscopic repair of cerebrospinal fluid rhinorrhea has been accepted by many otolaryngologists because it is minimally invasive, safe, quick, and accurate [4–7]. The repair methods include overlay and underlay techniques, and ‘‘button’’ and ‘‘bath-plug’’ closures [8,9]. The most commonly used repair materials are nasal mucosa, periosteum, fascia lata, muscle, fat, and artificial materials [10–13]. The success rates of transnasal endoscopic repair range between 76% and 97% [14]. Cerebrospinal fluid rhinorrhea can be definitively diagnosed by nasal endoscopy, determining the glucose content, biochemically measuring B2-transferrin, and by using computed tomography (CT) and magnetic resonance imagining (MRI) in most patients. The leak locations may be estimated by the cerebrospinal fluid outflow during surgery. In some patients, however, it is difficult to determine the leak locations because there may be less rhinorrhea, inflammatory swelling around the leak, or obstruction of the granulation tissue. It is a critical step of successful surgery to find leak locations during operations. For transnasal endoscopic repair procedure to be successful, it is critical to identify the leak locations during surgery. In this study, we used a combination of methods to detect the leak locations, including preoperative nasal endoscopy, radioisotope scanning, the intraoperative navigation system, and intraspinal saline injection. Our aim was to more accurately determine the leak locations to improve operative success rates.
2. Materials and methods 2.1. Subjects All study methods were approved by the Clinical Investigation Committee of the Xi Jing Hospital, Fourth Military Medical University. Between August 2001 and January 2006, 39 patients, were diagnosed with cerebrospinal fluid rhinorrhea by preoperative nasal endoscopy, radioisotope scanning, and glucose content determinations. The patients underwent transnasal endoscopic repair of cerebrospinal fluid rhinorrhea in our hospital. The patients ranged in age from 9 to 66 years (mean age of 36 years). Twenty patients were men and 19 were women. The locations of the leaks were determined prior to the operation in 27 cases. For the remaining 12 cases, we used a navigation system and intraspinal saline injection to find the leak locations during surgery. The patients’ general characteristics are shown in Table 1.
Table 1 All patient’s general condition. Items
Results (n = 39)
Age
Mean: 36 years Range: 9–66 years
Gender
Male: 20 Female: 19
Etiology of leak
Sinus surgery: 3 Trauma: 8 Spontaneous: 26 Skull base surgery 2
Duration of leak before surgery
Mean: 3.2 years Median: 2.5 years Range: 0–11 years
be determined by nasal endoscopy, preoperative radioisotope scanning was performed by injecting 99mTc-DTPA (diethylenetriaminepentaacetic acid) into the spinal subarachnoid space. Once the 99mTc-DTPA entered the cerebrospinal fluid pathway, 99mTc-DTPA scintigraphy was performed to determine whether cerebrospinal fluid rhinorrhea was present. 2.2.2. Navigation system One or two days prior to the operation, axial spiral CT was used to scan the nasal sinuses. The CT data were stored in the image-guided surgical system of the German Brain Lab for further use. Following anesthesia, patients were fitted with a head frame fixed with an infrared reflection sphere. The Z-Touch laser registration method was used, which had a mean error of 0.8 mm (0.5–1.2 mm). In all cases, the contouring of the ethmoid sinus was performed under image guidance. Cells in the ethmoid sinus were removed to expose the entire ethmoid roof. Alternatively, the lateral wall of the sphenoidal sinus was exposed by opening the sphenoidal sinus from its natural ostium. For the patients whose leak locations was undetected, intraspinal saline injection was then used to find the leak locations.
2.2. Methods
2.2.3. Intraspinal normal saline injection All patients lay on their sides and flexing their knee. A lumbar puncture was performed, and a spinal catheter that was connected to a pressure transducer (SONY Co.) was fixed. The cerebrospinal fluid pressure was determined before the operation. For the patients with leak locations that could not be determined by opening the sphenoidal or ethmoid sinuses, normal saline (in 20 ml volumes) was injected into the vertebral canal to observe the outflow of cerebrospinal fluid. When each 20-ml volume of normal saline was injected, the alteration of cerebrospinal fluid pressure was assessed using the pressure transducer to prevent a rapid increase in cerebrospinal fluid pressure.
2.2.1. Radioisotope scanning All cases (n = 39) first underwent nasal endoscopy to determine the leak locations. If the leak locations could not
2.2.4. Repairing leaks Small defects (<5 mm) were repaired with autologous muscle from the thigh using nasal endoscopy, but larger
L. Qiao et al. / Auris Nasus Larynx 38 (2011) 335–339
skull-base defects or skull-base developmental malformations were repaired using the intracranial-nasal approach [18]. 2.2.5. Postsurgical care Patients were positioned in a semi-reclining position and were intravenously administered 25% mannite (125 ml) and furosemide (20 mg) twice daily for 7–10 days. The iodoform gauze was drawn out of the nasal cavity 7 days following surgery, and the patients were discharged 10 days after the operation.
3. Results Of the 39 cases, the leak locations were determined by nasal endoscopy in 27 cases, but could not be determined for the remaining 12 cases. Twelve patients whose leak locations could not be determined by preoperative nasal endoscopy and radioisotope scanning underwent endoscopic repair of the cerebrospinal fluid leaks with the assistance of the image-guided system. The leak locations were suspected to be in the sphenoidal sinus in 3 cases. In one of these cases, after removing the granulation tissue, a small fissure was present on the posterior and superior wall of the sphenoidal sinus, and the image-guided system suggested that the fissure communicated with intracranial space. In the other 2 cases, under image guidance, after the edema mucosa was gently teased up, and 20 ml of saline was injected into vertebral canal, there was outflow present on lateral wall of the sphenoidal sinus. In 9 cases, the leaks were suspected to be located in the ethmoid sinus. After contouring the ethmoid sinus, the leak locations were well exposed in 3 cases. In the other 6 cases, there was no outflow of cerebrospinal fluid after contouring the ethmoid sinus. When these patients lay in the head-down tilt position and intracranial pressure was increased by pressing the internal jugular vein, fluid was still not present. Lastly, normal saline was injected into the vertebral canal (20 ml in 3 cases, 28 ml in 1 case, and 35 ml in 2 cases). Outflow was observed from a small fissure in the middle of the ethmoid roof in 4 cases. In 1 patient with substantial granulation tissue in the meninges, the leak location was in the posterior ethmoid sinus and close to the sphenoidal sinus. In the last case, the fluid was leaking from the cribriform plate, and the image-guided system suggested that the patient had a bony fissure defect. Details are shown in Tables 2 and 3. Follow-ups were performed 3–56 months (average 46 months) after surgery. Of the 39 cases, the first operation was successful in 36 cases, and leakage recurred in 3 cases.
337
Table 2 Leak locations as detected by different methods before the operations. Methods
Items
Results
Endoscope (n = 39)
Cerebrospinal fluid rhinorrhea
Yes: 27
Side of leak
Location of leak
Isotope scanning (n = 12)
Cerebrospinal fluid rhinorrhea Side of leak
Location of leak
Preoperative radioisotope scanning is a simple, safe, and reliable diagnostic method for cerebrospinal fluid rhinor-
Yes: 12 Uncertain: 0 Left: 0 Right: 0 Both: 0 Unknown: 12 Fovea ethmoigails: 0 Cribriform plate: 0 Sphenoid: 0 Frontal recess: 0 Unknown: 12
rhea. In patients where cerebrospinal fluid rhinorrhea is highly suspected, preoperative radioisotope scanning may be used to make a definitive diagnosis. The disadvantage of using preoperative radioisotope scanning is that the technique cannot be used to determine leak locations. The image-guided surgical system creates three-dimensional computerized reconstructions from preoperative CT and MRI images. Using this system, the surgeon can view the precise locations of surgical appliances in the surgical Table 3 Leak locations as detected by different methods during the operations. Methods
Items
Results
Image guidance (n = 4)
Side of leak
Left: 1
Location of leak
Intraspinal injection saline (n = 8)
Side of leak
Location of leak
4. Discussion
Uncertain: 12 Left: 7 Right: 20 Both: 0 Unknown: 12 Fovea ethmoigails: 17 Cribriform plate: 0 Sphenoid: 9 Frontal recess: 1 Unknown: 12
Right: 3 Both: 0 Unknown: 0 Fovea ethmoigails: 2 Cribriform plate: 1 Sphenoid: 1 Frontal recess: 0 Unknown: 0 Left: 3 Right: 5 Both: 0 Unknown: 0 Fovea ethmoigails: 5 Cribriform plate: 1 Sphenoid: 2 Frontal recess: 0 Unknown: 0
338
L. Qiao et al. / Auris Nasus Larynx 38 (2011) 335–339
field using the computer’s locating system. This surgical system has been widely used to perform a variety of nasal sinus and anterior skull base surgeries [15,16]. The imageguided system has the following advantages for repairing cerebrospinal fluid leaks. (1) It can be used to precisely estimate anatomical structures (particularly useful for the anterior wall and lateral wall of the sphenoidal sinus), and avoid the internal carotid artery and optic nerve. (2) After complete contouring of the ethmoid sinus by thoroughly removing cells on the ethmoid roof, it can detect leak locations easily. (3) The system can be used to visualize bony defects that cannot be viewed in the preoperative CT examination, which makes surgeries more accurate. Repair of cerebrospinal fluid leaks in the sphenoidal sinus is difficult, especially when they are located on the lateral wall because the surgical vision field is poor and the lateral wall of the sphenoidal sinus is adjacent to important structures, including the internal carotid artery and optic nerve. We used the image-guided surgical system to accurately localize the anterior and lateral wall of the sphenoidal sinus. Additionally, the system allowed the optic nerve and internal carotid artery to be visualized so the surgeon could avoid these important structures. The visualization of bony defects in the three-dimensional reconstruction increased surgical safety and the accuracy of localizing leaks, which improved the surgeons’ confidence. The leaks were accurately localized using the above approaches in our 39 cases, which improved the success rate of the surgeries. Simple nasal endoscopic surgery is not suitable for larger skull-base defects or skull-base malformations. Of 39 patients, 2 cases underwent intracranial-nasal repair (both with spontaneous cerebrospinal fluid leaks and meningoencephaloceles, one case had a 0.8 cm 1.4 cm skull-base defect and the other had a skull-base conical malformation), and the first repair was successful. Three patients experienced recurrence of the cerebrospinal fluid leaks, and all three were treated by reoperation. In one of the recurrent cases, although the leak was repaired by nasal endoscopy, when the intracranial pressure was increased, the cerebrospinal fluid leak reoccurred because of multiple bony defects on the cribriform plate. A second intracranial repair was successfully performed to treat the patient. Therefore, it is critical to determine the leak locations and select the appropriate surgical procedures to improve the success rates for nasal endoscopic repair of cerebrospinal fluid leaks. To repair traumatic or spontaneous cerebrospinal fluid leaks, one of the greatest challenges is detecting the precise leak location. Generally, CT and MRI are not very useful for determining the leak location. Intraoperative outflow of cerebrospinal fluid is decreased and sometimes is not present due to the patients’ position when they are lying down, which makes it difficult to determine the leak location. When the leak locations cannot be detected, most surgeons inject fluorescein into the vertebral canal to observe the
outflow of fluorescein though the nasal endoscope [17]. This method is indeed practicable, but the high concentration of fluorescein may contribute to nerve injury. We injected saline into the vertebral canal to increase cerebrospinal fluid pressure, which resulted in a visible outflow of cerebrospinal fluid, which could be clearly observed under the nasal endoscope. During the endoscopic repair of the cerebrospinal fluid leak, we injected 15–35 ml of normal saline into the vertebral canal in 8 cases, and all of the leak locations were well exposed. Monitoring cerebrospinal fluid pressure during surgery is important because most patients with spontaneous cerebrospinal fluid leaks also have high cerebrospinal fluid pressure.
5. Conclusion For cerebrospinal fluid rhinorrhea patients whose leak locations are difficult to determine, surgeons can better determine the orientation of the surgical fields by performing radioisotope scanning and intraspinal saline injections and by using image-guided surgical systems. These safe and effective methods can be used to successfully detect leak locations during transnasal endoscopic repair of cerebrospinal fluid leaks.
References [1] Kelley TF, Stankiewicz JA, Chow JM, Origitano TC, Shea J. Endoscopic closure of postsurgical anterior cranial fossa cerebrospinal fluid leaks. Neurosurgery 1996;39:743–6. [2] Lopatin AS, Kapitanov DN, Potapov AA. Endonasal endoscopic repair of spontaneous cerebrospinal fluid leaks. Arch Otolaryngol Head Neck Surg 2003;129:859–63. [3] Carrau RL, Snyderman CH, Kassam AB. The management of cerebrospinal fluid leaks in patients at risk for high-pressure hydrocephalus. Laryngoscope 2005;115:205–12. [4] Briggs RJ, Wormald PJ. Endoscopic transnasal intradural repair of anterior skull base cerebrospinal fluid fistulae. J Clin Neurosci 2004;11:597–9. [5] Bernal-Sprekelsen M, Alobid I, Mullol J, Trobat F, Toma´s-Barbera´n M. Closure of cerebrospinal fluid leaks prevents ascending bacterial meningitis. Rhinology 2005;43:277–81. [6] McMains KC, Gross CW, Kountakis SE. Endoscopic management of cerebrospinal fluid rhinorrhea. Laryngoscope 2004;114:1833–7. [7] Lee TJ, Huang CC, Chuang CC, Huang SF. Transnasal endoscopic repair of cerebrospinal fluid rhinorrhea and skull base defect: ten-year experience. Laryngoscope 2004;114:1475–81. [8] El-Banhawy OA, Halaka AN, El-Hafiz Shehab El-Dien A, Ayad H. Subcranial transnasal repair of cerebrospinal fluid rhinorrhea with free autologous grafts by the combined overlay and underlay techniques. Minim Invasive Neurosurg 2004;47:197–202. [9] Wormald PJ, McDonogh M. The bath-plug closure of anterior skull base cerebrospinal fluid leaks. Am J Rhinol 2003;17:299–305. [10] Lorenz RR, Dean RL, Hurley DB, Chuang J, Citardi MJ. Endoscopic reconstruction of anterior and middle cranial fossa defects using acellular dermal allograft. Laryngoscope 2003;113:496–501. [11] Weber SM, Kim J, Delashaw JB, Wax MK. Radial forearm free tissue transfer in the management of persistent cerebrospinal fluid leaks. Laryngoscope 2005;115:968–72.
L. Qiao et al. / Auris Nasus Larynx 38 (2011) 335–339 [12] Costantino PD, Hiltzik DH, Sen C, Friedman CD, Kveton JF, Snyderman CF, et al. Sphenoethmoid cerebrospinal fluid leak repair with hydroxyapatite cement. Arch Otolaryngol Head Neck Surg 2001; 127:588–93. [13] Kelly DF, Oskouian RJ, Fineman I. Collagen sponge repair of small cerebrospinal fluid leaks obviates tissue grafts and cerebrospinal fluid diversion after pituitary surgery. Neurosurgery 2001;49: 885–90. [14] Ismail AS, Costantino PD, Sen C. Transnasal transsphenoidal endoscopic repair of CSF leakage using multilayer acellular dermis. Skull Base 2007;17:125–32.
339
[15] Wise SK, DelGaudio JM. Computer-aided surgery of the paranasal sinuses and skull base. Expert Rev Med Devices 2005;2:395–408. [16] Hemmerdinger SA, Jacobs JB, Lebowitz RA. Accuracy and cost analysis of image-guided sinus surgery. Otolaryngol Clin North Am 2005;38:453–60. [17] Stammberger H, Greistorfer K, Wolf G, et al. Surgical occlusion of cerebrospinal fistulas of the anterior skull base using intrathecal sodium fluorescein. Laryngorhinootologie 1997;76:595–607. [18] Woodworth BA, Schlosser RJ, Faust RA, Bolger WE. Evolutions in the management of congenital intranasal skull base defects. Arch Otolaryngol Head Neck Surg 2004;130:1283–8.