- Email: [email protected]
Surgical Outcomes of Revision Microvascular Decompression for Persistent or Recurrent Hemifacial Spasm After Surgery: Analysis of Radiologic and Intraoperative Findings
Original Article
Surgical Outcomes of Revision Microvascular Decompression for Persistent or Recurrent Hemifacial Spasm After Surgery: Analysis of Radiologic and Intraoperative Findings Chang Kyu Park, Seung Hwan Lee, Bong Jin Park
OBJECTIVES: Microvascular decompression (MVD) is the treatment of choice for hemifacial spasm (HFS), due to the high rate of complete resolution associated with MVD. However, some patients experience recurrent or persistent symptoms after surgery. In this study, we evaluated the causes of recurrence or failure based on our surgical experience with revision of MVD for HFS and analyzed the relationship between surgical outcomes and radiologic and intraoperative findings.
-
METHODS: Among more than 2500 patients who underwent MVD surgery for HFS, 23 patients received a second MVD in our hospital from January 2002 to December 2017. Three-dimensional time-of-flight magnetic resonance angiography and reconstructed imaging were used to identify the culprit vessel and its conflict on the root exit zone (REZ) of the facial nerve. We reviewed patients’ medical records and operation videos to identify the missing points of first surgery.
and Teflon malposition patient groups. There were no clear neurovascular contacts in the patients with severe Teflon adhesion. CONCLUSIONS: The decision regarding secondary MVD for persistent or recurrent spasm is troubling. However, if neurovascular contact was observed in the MRI of the patient and there were offending vessels, the surgical outcome might be favorable.
-
-
RESULTS: In our experience with revision of MVD, 8 patients had incomplete decompression, such as singlevessel decompression of multiple offending vessels. Teflon was not detected at the REZ but was found in other locations in 12 patients. Three patients had severe adhesion with previous Teflon around the REZ. Nineteen patients had excellent surgical outcomes at immediate postoperative evaluation; 20 patients showed spasm disappearance at 1 year after surgery and 3 patients showed persistent symptoms. Neurovascular contacts around the REZ of the facial nerve were revealed on magnetic resonance imaging of incomplete decompression
-
Key words Failed surgery - Hemifacial spasm - Microvascular decompression - Recurrent hemifacial spasm - Teflon granuloma -
Abbreviations and Acronyms HFS: Hemifacial spasm MRI: Magnetic resonance imaging MVD: Microvascular decompression REZ: Root entry zone
WORLD NEUROSURGERY -: e1-e6, - 2019
INTRODUCTION
H
emifacial spasm (HFS) is a benign, painless, stressful condition. Patients characteristically seek treatment only when socioeconomic activities are extremely bothered by its symptoms. Jannetta et al1 have described microvascular decompression (MVD) surgery as the most effective treatment for HFS, with success rates of 86%92%.2-5 However, despite successful decompression, there have been several instances of remaining spasms or recurrent spasms several days after surgery. Because these may constitute delayed effects of the surgery, the effect of surgery cannot be judged only on immediate status after surgery.6 Hence multiple studies regarding the causes and treatments of remaining or recurrent spasms are ongoing. To date, on the basis of intraoperative findings, a few authors have suggested underlying causes of remnant or recurrent spasms as follows: incomplete decompression at the time of initial surgery and/or newly originated vascular compression. Although it is easy to say that addressing those factors correct in the revision surgery, there could be numerous difficulties in making decisions for the revision surgery with an additional hazardous
Department of Neurosurgery, Kyung Hee University College of Medicine, Seoul, Korea To whom correspondence should be addressed: Bong Jin Park, M.D., Ph.D. [E-mail: [email protected]] Citation: World Neurosurg. (2019). https://doi.org/10.1016/j.wneu.2019.07.191 Journal homepage: www.journals.elsevier.com/world-neurosurgery Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2019 Elsevier Inc. All rights reserved.
www.journals.elsevier.com/world-neurosurgery
e1
ORIGINAL ARTICLE CHANG KYU PARK ET AL.
REVISION MVD FOR HEMIFACIAL SPASM
complication, doing the decompression under adhesions, and assessing the complete decompression of the facial nerve.7 In previous studies of failed MVD surgery, the causes of the failure were Teflon-related adhesion, Teflon granuloma, insufficient Teflon material, Teflon malposition, or new culprit vessels.8-11 In this study, we classified the causes of various failures identified at the time of surgery into 3 categories and investigated the relationship between these causes and surgical outcomes. In addition, we attempted to suggest the appropriate indications for secondary MVD by analyzing the association between secondary preoperative imaging, surgical findings, and treatment outcomes.
recurrent symptoms. All patients were followed up for at least 1 year postoperatively.
Intraoperative Findings Intraoperative findings during revision MVD surgeries were analyzed. Three general scenarios were noted in all cases: 1) inappropriate placement of Teflon pellets (Teflon malposition); 2) incomplete decompression; and 3) severe adhesion and granulation around previously inserted Teflon (Teflon granuloma). In 12 patients, inappropriate placement of Teflon pellets was noted, at locations other than the root entry zone (REZ). Among these 12 patients in the Teflon malposition group, 1 had Teflon pellets inserVideo available at MATERIALS AND METHODS www.sciencedirect.com ted around cranial nerve IX (Video 1), 1 had Teflon pellets inserted at the internal auditory meatus Patient Population (IAM), 5 had no Teflon pellets around the REZ, and 5 had This retrospective study included approximately 2500 patients who Teflon pellets near cranial nerve VII but not at the REZ. All of underwent MVD surgery for HFS in our Cranial Nerve Disorders patients with Teflon malposition had persistent symptoms after Center from January 2002 to December 2017. Among them, 23 initial surgery. patients underwent revision MVD for remaining or recurrent HFS Eight patients demonstrated incomplete decompression. In 2 of after the initial surgery. Medical records including admission note, these 8 patients, Teflon pellets were found at the REZ; however, radiologic reports, intraoperative videos, and surgical notes of because of insufficient decompression, vascular compression those patients were reviewed to analyze the patient data. remained. In 3 other patients, portions of offending vessels were in contact with the REZ (Video 2); in 2 patients, distal portions of Criteria for Revision Microvascular Decompression Surgery and cranial nerve VII were in contact with the offending vessels. Surgical Procedures Moreover, in the remaining 1 patient, the offending vessel was For indications of revision MVD, patients were divided into 2 mistakenly identified as the posterior inferior cerebellar artery groups: persistent spasm and recurrent spasm. The persistent during the initial surgery, whereas the actual culprit vessel was spasm group included patients with persistent symptoms until 1 the anterior inferior cerebellar artery. Especially the patients year after surgery, and the recurrent spasm group consisted of with distal part compression and those with perforator offender patients who had a symptom-free period for more than 1 year after had good results after initial surgery; however, they suffered surgery but who subsequently developed symptoms. All patients from recurrent symptoms. underwent preoperative magnetic resonance imaging (MRI) In the other 3 patients, Teflon pellets were found strongly including 3-dimensional time-of-flight magnetic resonance angiadhered to the offending vessels. Three patients showed recurrent ography and reconstructed imaging to confirm neurovascular symptoms. The isolation of Teflon pellets and offending vessels contact. In addition, patients underwent facial electromyography was performed carefully in 2 of these 3 patients; in 1 patient, such to assess whether they showed an abnormal muscle response. isolation could not be achieved. The intraoperative findings are 12 Surgeries were performed as previously described. In particular, summarized in Table 1. muscle dissections were performed carefully for patients who had not undergone cranioplasty. Surgical Outcomes All patients who underwent revision MVD surgeries were followed Evaluation After Revision Surgery up for 1 year. Nineteen patients experienced immediate relief of Patients’ postoperative symptoms were categorized into 2 groups: symptoms after secondary surgery. Twenty patients (86%) relief, no symptoms or intermittent symptoms without discomdemonstrated symptom relief at 1 year after secondary surgery. fort; and no relief, no improvement or symptom with discomfort. There was 1 complication, permanent facial palsy of HousePatients were reevaluated at 1 year postoperatively. Brackman grade III, in 1 patient. In most patients with Teflon malposition, symptoms were improved immediately postoperatively; all had excellent outcomes at 1 year after secondary RESULTS surgery. Among patients with incomplete decompression, 5 Patients’ Characteristics experienced symptomatic relief immediately postoperatively. In Of the 23 patients in this analysis, 8 were men and 15 were women. addition, after 1 year of follow-up, 7 patients had improved to Two of the 23 patients had a history of MVD surgery in our center; symptomatic relief. One patient with distal compression had no the remaining 21 patients had undergone MVD procedures in symptomatic relief at 1 year postoperatively. Finally, among paother hospitals. The mean age of the patients was 55.3 years tients with Teflon granuloma, improvements were noted in only 1 (range: 33e70 years), and the mean duration between initial MVD who had undergone removal of granuloma, at the time of 1-year surgery and revision surgery was 44.9 months. Sixteen patients follow-up visits (Table 2). Twenty-two patients showed abnormal had persistent symptoms after initial surgery, and 7 patients had muscle responses on preoperative facial electromyography; this
e2
www.SCIENCEDIRECT.com
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.07.191
ORIGINAL ARTICLE CHANG KYU PARK ET AL.
REVISION MVD FOR HEMIFACIAL SPASM
Table 1. Characteristics of Intraoperative Findings Intraoperative Finding
Table 2. Surgical Outcomes of Reoperation Immediately After Second MVD
Patients, n
1 Year After Second MVD
Teflon malposition Absent from REZ
5
CN IX
1
Intraoperative Finding
IAM
1
Teflon malposition
Vicinity of CN VII
5
Absent from REZ
5
5
0
5
0
Prior offender
2
CN IX
1
1
0
1
0
Perforators
3
IAM
1
1
0
1
0
2
Vicinity of CN VII
5
5
0
5
0
2
0
2
0
Incomplete decompression
Compression of distal portion New offender
1
Strong adhesion
3
n, number; REZ, root exit zone; CN, cranial nerve; IAM, internal acoustic meatus.
abnormal muscle response disappeared after the second MVD in 18 patients. Postoperatively, 19 of 20 patients with symptomatic relief showed no abnormal muscle response. However, 3 patients with persistent symptoms showed persistent abnormal muscle response. There was a statistical significance between the absence of symptoms and absence of abnormal muscle response (P ¼ 0.04). Relationship Among Preoperative Magnetic Resonance Imaging, Intraoperative Findings, and Surgical Outcomes On MRI of the Teflon malposition and incomplete decompression groups, neurovascular contacts that may cause symptoms were revealed near the REZ of the facial nerve (Figure 1). However, the patients with Teflon granulations did not have a clear neurovascular contact around the REZ (Figure 2). In most cases, prior Teflon was observed as a lesion with low signal intensity on T1-and T2-weighted images. Even in patients with granulation, no enhancement around the Teflon was observed. The presence of neurovascular contact in preoperative MRI was statistically significant for the surgical outcome (P ¼ 0.03), and the outcome of surgery was favorable when neurovascular contact was evident. DISCUSSION Recurrence and lack of symptomatic improvement comprise unwanted complications after MVD surgeries, such that the procedure appears ineffective. In addition, this increases confusion among patients who had previously undergone MVD surgery and experience recurrences. Although a few patients experience recurrences or show no symptomatic improvement despite surgery, the importance of MVD surgery for relieving HFS remains unquestionable and does not signify failure of the procedure. Goto et al6 initially described the delayed effect of the MVD procedure and suggested that follow-up should be performed for at least 1 year postoperatively to confirm the outcome.
WORLD NEUROSURGERY -: e1-e6, - 2019
Patients, Symptom Symptom n Relief Persistent Relief Persistent
Incomplete decompression Prior offender
2
Perforators
3
2
1
3
0
Compression of distal portion
2
0
2
1
1
New offender
1
1
0
1
0
Strong adhesion
3
1
2
1
2
n, number; MVD, microvascular decompression; REZ, root exit zone; CN, cranial nerve; IAM, internal acoustic meatus.
Similarly, Jo et al13 stated that the natural course after MVD surgery is sufficiently diverse that it is difficult to determine the optimal time for determining the failure or delayed effect of MVD surgery. Hence in this study, we included patients who underwent revision MVD surgeries for recurrences or lack of symptomatic improvement over 1 year of follow-up to assess delayed effects. In our study, we noted several potential sources of symptoms and observed significant differences between causes that led to persistent spasms and those that led to recurrent spasms. In patients who exhibited intraoperative malposition of Teflon pellets, persistent spasms after initial surgery could be attributed to incomplete or insufficient decompression of the offending vessels. Similarly, in patients with incomplete intraoperative findings, decompression had been performed using Teflon pellets; however, due to inadequate decompression, the offending vessels continuously excited the REZ, which caused persistent spasms postoperatively. In support of this theory, Yuan et al14 analyzed surgical outcomes in 23 patients who underwent revision MVD surgeries; they concluded that the incomplete decompression of the offending vessel was the main source of persistent symptoms after MVD surgery. Moreover, they categorized factors underlying incomplete decompression, which included inaccurate identification of offending vessels, inaccurate insertion of Teflon pellets, and insertion of inadequate volumes of decompressants. Similarly, Li et al15 described the findings in 1 patient with persistent symptoms after surgery: during revision MVD, other branches of the posterior inferior cerebellar artery
www.journals.elsevier.com/world-neurosurgery
e3
ORIGINAL ARTICLE CHANG KYU PARK ET AL.
REVISION MVD FOR HEMIFACIAL SPASM
Figure 1. Magnetic resonance imaging of the patient with Teflon malposition. (A) 3-Dimensional time-of-flight image showed neurovascular contact to the root entry zone of the right facial nerve (white arrow). (B)
were confirmed to compress the REZ. This may have been an underlying cause of persistent symptoms for patients in our study, although all underwent initial MVD procedures. In addition, all patients mentioned earlier showed recurrent symptoms after initial surgeries. The cause of recurrence was
Figure 2. Magnetic resonance imaging of the patient with Teflon granulation. (A) There was no neurovascular contact around the root entry zone (REZ) of left facial nerve (white arrow). (B) Prior Teflon with dark signal
e4
www.SCIENCEDIRECT.com
Prior Teflon with low signal density was revealed around IX and X cranial nerves (white asterisk). (C) T2-weighted image, Teflon was showed in same site (white asterisk).
either a new offending vessel or a new vascular contact point. Similarly, some patients initially showed disappearance of symptoms; however, these then recurred, combined with strong adhesion. Thus Li et al15 proposed that adhesion of Teflon pellets to the facial nerve acted in a manner similar to that of an offending
intensity was revealed in REZ of facial nerve (white arrowhead). A slight contrast enhancement was shown. (C) Dark signal intensity Teflon was shown in the same site.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.07.191
ORIGINAL ARTICLE CHANG KYU PARK ET AL.
vessel, thereby stimulating the nerve itself. Park et al9 suggested that the adhesion between decompressants and the nerve may contribute to recurrent HFS. In the present study, we were able to release the adhesion while avoiding nerve injury. As a result, patients who underwent removal of Teflon granulomas achieved favorite outcomes. Moreover, we confirmed that the Teflon granulation was the source of symptoms in patients who underwent MVD because we found facial nerve indentation after removal of Teflon granuloma. However, the circumstances under which granulation occurs remain unclear. There is no definite study about it. However, histologic studies of Teflon granulomas showed that their features were primarily associated with inflammation.11,16 Chen et al17 proposed a possible pathogenesis underlying the onset of Teflon granulomas: When Teflon contacts the dura or tentorium, a foreign body reaction is stimulated, comprising an initial inflammatory giant cell response. This leads to increased blood supply, followed by inflammatory reactions and development of granulation. In the present study, we analyzed the presence of Teflon in contact with the dura, as well as the presence of a vessel-rich environment around Teflon pellets in patients with Teflon granulomas; however, we found no definite association due to the small sample size. Therefore to prevent development of Teflon granulomas, it is recommended that small vessels be treated carefully to prevent bleeding and not to insert too much Teflon. The current study enrolled patients whose symptoms persisted for >1 year after initial surgery, and in whom we were able to attain effective results. However, if revision MVD is needed, there is no ideal timing for the procedure. Zhong et al18 reported success when revision MVD surgeries were performed early in patients with a history of failed initial MVD procedures; conversely, Wang et al19 obtained good results when performing delayed revision MVDs. Hence reviews of surgical outcomes by several authors suggest that the exact timing of revision MVD surgery is not particularly important; in contrast, the important aspect remains selection of patients who are most likely to benefit from the procedure. Park et al9 also performed revision MVD surgeries only in patients whose symptoms did not disappear after initial surgery, with offending vessels that could be confirmed in 3-dimensional time-of-flight magnetic resonance imaging. They reported considerable success following revision surgeries. Conversely, Yamaki et al20 performed revision MVD surgeries in 10 patients whose symptoms persisted or recurred after surgery; they reported that no key offending vessels could be identified intraoperatively and that surgical outcomes remained poor. In the present study, we performed revision MVDs in patients in whom clear or doubtful offending vessels could be identified in radiologic examinations; thus we were able to achieve good surgical outcomes to patients with clear neurovascular contact. Therefore decisions regarding revision MVD should be based on the presence of clinically specific symptoms, as well as on the evidence of offending vessels in radiologic examinations; this facilitates achievement of good surgical outcomes from the revision procedures. As mentioned previously, the identification of offending vessels in radiologic examinations forms the basis for revision surgery.
WORLD NEUROSURGERY -: e1-e6, - 2019
REVISION MVD FOR HEMIFACIAL SPASM
Despite this basis for surgery, intraoperative findings could be diverse. In the present study, we attempted to categorize intraoperative findings that were ultimately related to surgical outcomes. In patients in the first category (malpositioned Teflon pellets), no proper decompression of the REZ could be performed; hence if proper decompression can be performed, good surgical outcomes can be achieved. Similarly, in patients in the second category (incomplete decompression or compression from additional vessels), the cause of the spasm could be confirmed; if it was adequately and properly decompressed, good results could be achieved. However, when strong adhesion was observed between the nerve root and Teflon pellets, careful consideration was needed regarding continuation of the decompression procedure. In this situation, the adhesion itself can mimic the offending vessel, and relieving the adhesion can facilitate decompression and good results; however, extensive dissection to relieve the adhesion could also lead to injuries to the nerve, blood vessel, or both. These injuries can ultimately only result in unfavorable outcomes; thus proper visualization and understanding of the adhesion are essential prior to continuation of surgery. In the present study, strong adhesions were noted in 3 patients; no further surgical intervention was performed in 1 of these patients. Notably, no symptomatic recovery could be expected in this patient. In summary, good surgical outcomes can be achieved if adequate decompressive procedures are performed when Teflon pellets are not inserted at the REZ or in its general vicinity. These patients have a clear neurovascular contact on preoperative MRI. In patients with severe Teflon adhesion, there is no clear neurovascular contact around the REZ in the preoperative MRI, although severe adhesion or granulation may be present even if no enhancement is observed. In patients with these imaging characteristics, the success rate of revision MVD may be lowered. A considerable limitation of this study is that there were no previous operative records regarding initial surgeries performed at external hospitals; thus we were unable to confirm the types of decompressants used during initial surgeries, particularly in patients with intraoperative nonidentifiable decompressing agents. If proper identification could have been made of the types of decompressants used in the initial surgeries, intraoperative approaches might have been adjusted accordingly.
CONCLUSIONS In patients who have undergone MVD surgeries for HFS but persistent or recurrent symptoms remain, revision MVD procedures can be helpful if the symptoms can be clearly discerned both clinically and radiologically. Good surgical outcomes can be achieved if adequate decompression is performed in patients who have no Teflon pellets at the REZ or in its general vicinity, or if inadequate and improper decompressions were performed involving nonoffending blood vessels. And if a clear neurovascular contact is observed in the preoperative MRI, a favorable surgical outcome can be predicted.
www.journals.elsevier.com/world-neurosurgery
e5
ORIGINAL ARTICLE CHANG KYU PARK ET AL.
REFERENCES 1. Jannetta PJ, Abbasy M, Maroon JC, et al. Etiology and definitive microsurgical treatment of hemifacial spasm. Operative techniques and results in 47 patients. J Neurosurg. 1977;47:321-328. 2. Ma Q, Zhang W, Li G, et al. Analysis of therapeutic effect of microvascular decompression surgery on idiopathic hemifacial spasm. J Craniofac Surg. 2014;25:1810-1813. 3. Xia L, Zhong J, Zhu J, et al. Delayed relief of hemifacial spasm after microvascular decompression. J Craniofac Surg. 2015;26:408-410. 4. Zhong J, Zhu J, Sun H, et al. Microvascular decompression surgery: surgical principles and technical nuances based on 4000 cases. Neurol Res. 2014;36:882-893. 5. Barker FG 2nd, Jannetta PJ, Bissonette DJ, et al. Microvascular decompression for hemifacial spasm. J Neurosurg. 1995;82:201-210. 6. Goto Y, Matsushima T, Natori Y, et al. Delayed effects of the microvascular decompression on hemifacial spasm: a retrospective study of 131 consecutive operated cases. Neurol Res. 2002;24: 296-300. 7. Kureshi SA, Wilkins RH. Posterior fossa reexploration for persistent or recurrent trigeminal neuralgia or hemifacial spasm: surgical findings and therapeutic implications. Neurosurgery. 1998;43: 1111-1117. 8. Ravina K, Strickland BA, Rennert RC, et al. Revision microvascular decompression for
e6
www.SCIENCEDIRECT.com
REVISION MVD FOR HEMIFACIAL SPASM
trigeminal neuralgia and hemifacial spasm: factors associated with surgical failure. J Neurol Surg B Skull Base. 2019;80:31-39.
16. Deep NL, Graffeo CS, Copeland WR 3rd, et al. Teflon granulomas mimicking cerebellopontine angle tumors following microvascular decompression. Laryngoscope. 2017;127:715-719.
9. Park YS, Chang JH, Cho J, et al. Reoperation for persistent or recurrent hemifacial spasm after microvascular decompression. Neurosurgery. 2006; 58:1162-1167.
17. Chen J, Lee S, Lui T, et al. Teflon granuloma after microvascular decompression for trigeminal neuralgia. Surg Neurol. 2000;53:281-287.
10. Fan ZM, Han YC, Li JF, et al. [Revision surgery of the posterior fossa cranial nerves disease]. Zhonghua Er Bi Yan Hou Tou Jing Wai Ke Za Zhi. 2009; 44:364-368.
18. Zhong J, Zhu J, Li ST, et al. An analysis of failed microvascular decompression in patients with hemifacial spasm: focused on the early reoperative findings. Acta Neurochir (Wien). 2010;152:2119-2123.
11. Oda K, Higuchi T, Murai Y, et al. Teflon granuloma after microvascular decompression for hemifacial spasm: a case report and literature review. Neurosurg Rev. 2017;40:513-516. 12. Kim JP, Park BJ, Choi SK, et al. Microvascular decompression for hemifacial spasm associated with vertebrobasilar artery. J Kor Neurosurg Soc. 2008;44:131-135. 13. Jo KW, Kong DS, Park K. Microvascular decompression for hemifacial spasm: long-term outcome and prognostic factors, with emphasis on delayed cure. Neurosurg Rev. 2013;36:297-301.
19. Wang X, Thirumala PD, Shah A, et al. Microvascular decompression for hemifacial spasm: focus on late reoperation. Neurosurg Rev. 2013;36:637-643. 20. Yamaki T, Hashi K, Niwa J, et al. Results of reoperation for failed microvascular decompression. Acta Neurochir (Wien). 1992;115:1-7.
Conflict of interest statement: This work was supported by a grant from Kyung Hee University in 2018 (KHU-20182176). The authors declare that the article content was composed in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Received 15 May 2019; accepted 26 July 2019
14. Yuan Y, Zhang L, Zhang SX, et al. [Causes of ineffectiveness of microvascular decompression for hemifacial spasm]. Zhonghua Wai Ke Za Zhi. 2003;41:362-364.
Citation: World Neurosurg. (2019). https://doi.org/10.1016/j.wneu.2019.07.191
15. Li CS. Varied patterns of postoperative course of disappearance of hemifacial spasm after microvascular decompression. Acta Neurochir (Wien). 2005;147:617-620.
Available online: www.sciencedirect.com
Journal homepage: www.journals.elsevier.com/worldneurosurgery 1878-8750/$ - see front matter ª 2019 Elsevier Inc. All rights reserved.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.07.191
Surgical Outcomes of Revision Microvascular Decompression for Persistent or Recurrent Hemifacial Spasm After Surgery: Analysis of Radiologic and Intraoperative Findings Chang Kyu Park, Seung Hwan Lee, Bong Jin Park
OBJECTIVES: Microvascular decompression (MVD) is the treatment of choice for hemifacial spasm (HFS), due to the high rate of complete resolution associated with MVD. However, some patients experience recurrent or persistent symptoms after surgery. In this study, we evaluated the causes of recurrence or failure based on our surgical experience with revision of MVD for HFS and analyzed the relationship between surgical outcomes and radiologic and intraoperative findings.
-
METHODS: Among more than 2500 patients who underwent MVD surgery for HFS, 23 patients received a second MVD in our hospital from January 2002 to December 2017. Three-dimensional time-of-flight magnetic resonance angiography and reconstructed imaging were used to identify the culprit vessel and its conflict on the root exit zone (REZ) of the facial nerve. We reviewed patients’ medical records and operation videos to identify the missing points of first surgery.
and Teflon malposition patient groups. There were no clear neurovascular contacts in the patients with severe Teflon adhesion. CONCLUSIONS: The decision regarding secondary MVD for persistent or recurrent spasm is troubling. However, if neurovascular contact was observed in the MRI of the patient and there were offending vessels, the surgical outcome might be favorable.
-
-
RESULTS: In our experience with revision of MVD, 8 patients had incomplete decompression, such as singlevessel decompression of multiple offending vessels. Teflon was not detected at the REZ but was found in other locations in 12 patients. Three patients had severe adhesion with previous Teflon around the REZ. Nineteen patients had excellent surgical outcomes at immediate postoperative evaluation; 20 patients showed spasm disappearance at 1 year after surgery and 3 patients showed persistent symptoms. Neurovascular contacts around the REZ of the facial nerve were revealed on magnetic resonance imaging of incomplete decompression
-
Key words Failed surgery - Hemifacial spasm - Microvascular decompression - Recurrent hemifacial spasm - Teflon granuloma -
Abbreviations and Acronyms HFS: Hemifacial spasm MRI: Magnetic resonance imaging MVD: Microvascular decompression REZ: Root entry zone
WORLD NEUROSURGERY -: e1-e6, - 2019
INTRODUCTION
H
emifacial spasm (HFS) is a benign, painless, stressful condition. Patients characteristically seek treatment only when socioeconomic activities are extremely bothered by its symptoms. Jannetta et al1 have described microvascular decompression (MVD) surgery as the most effective treatment for HFS, with success rates of 86%92%.2-5 However, despite successful decompression, there have been several instances of remaining spasms or recurrent spasms several days after surgery. Because these may constitute delayed effects of the surgery, the effect of surgery cannot be judged only on immediate status after surgery.6 Hence multiple studies regarding the causes and treatments of remaining or recurrent spasms are ongoing. To date, on the basis of intraoperative findings, a few authors have suggested underlying causes of remnant or recurrent spasms as follows: incomplete decompression at the time of initial surgery and/or newly originated vascular compression. Although it is easy to say that addressing those factors correct in the revision surgery, there could be numerous difficulties in making decisions for the revision surgery with an additional hazardous
Department of Neurosurgery, Kyung Hee University College of Medicine, Seoul, Korea To whom correspondence should be addressed: Bong Jin Park, M.D., Ph.D. [E-mail: [email protected]] Citation: World Neurosurg. (2019). https://doi.org/10.1016/j.wneu.2019.07.191 Journal homepage: www.journals.elsevier.com/world-neurosurgery Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2019 Elsevier Inc. All rights reserved.
www.journals.elsevier.com/world-neurosurgery
e1
ORIGINAL ARTICLE CHANG KYU PARK ET AL.
REVISION MVD FOR HEMIFACIAL SPASM
complication, doing the decompression under adhesions, and assessing the complete decompression of the facial nerve.7 In previous studies of failed MVD surgery, the causes of the failure were Teflon-related adhesion, Teflon granuloma, insufficient Teflon material, Teflon malposition, or new culprit vessels.8-11 In this study, we classified the causes of various failures identified at the time of surgery into 3 categories and investigated the relationship between these causes and surgical outcomes. In addition, we attempted to suggest the appropriate indications for secondary MVD by analyzing the association between secondary preoperative imaging, surgical findings, and treatment outcomes.
recurrent symptoms. All patients were followed up for at least 1 year postoperatively.
Intraoperative Findings Intraoperative findings during revision MVD surgeries were analyzed. Three general scenarios were noted in all cases: 1) inappropriate placement of Teflon pellets (Teflon malposition); 2) incomplete decompression; and 3) severe adhesion and granulation around previously inserted Teflon (Teflon granuloma). In 12 patients, inappropriate placement of Teflon pellets was noted, at locations other than the root entry zone (REZ). Among these 12 patients in the Teflon malposition group, 1 had Teflon pellets inserVideo available at MATERIALS AND METHODS www.sciencedirect.com ted around cranial nerve IX (Video 1), 1 had Teflon pellets inserted at the internal auditory meatus Patient Population (IAM), 5 had no Teflon pellets around the REZ, and 5 had This retrospective study included approximately 2500 patients who Teflon pellets near cranial nerve VII but not at the REZ. All of underwent MVD surgery for HFS in our Cranial Nerve Disorders patients with Teflon malposition had persistent symptoms after Center from January 2002 to December 2017. Among them, 23 initial surgery. patients underwent revision MVD for remaining or recurrent HFS Eight patients demonstrated incomplete decompression. In 2 of after the initial surgery. Medical records including admission note, these 8 patients, Teflon pellets were found at the REZ; however, radiologic reports, intraoperative videos, and surgical notes of because of insufficient decompression, vascular compression those patients were reviewed to analyze the patient data. remained. In 3 other patients, portions of offending vessels were in contact with the REZ (Video 2); in 2 patients, distal portions of Criteria for Revision Microvascular Decompression Surgery and cranial nerve VII were in contact with the offending vessels. Surgical Procedures Moreover, in the remaining 1 patient, the offending vessel was For indications of revision MVD, patients were divided into 2 mistakenly identified as the posterior inferior cerebellar artery groups: persistent spasm and recurrent spasm. The persistent during the initial surgery, whereas the actual culprit vessel was spasm group included patients with persistent symptoms until 1 the anterior inferior cerebellar artery. Especially the patients year after surgery, and the recurrent spasm group consisted of with distal part compression and those with perforator offender patients who had a symptom-free period for more than 1 year after had good results after initial surgery; however, they suffered surgery but who subsequently developed symptoms. All patients from recurrent symptoms. underwent preoperative magnetic resonance imaging (MRI) In the other 3 patients, Teflon pellets were found strongly including 3-dimensional time-of-flight magnetic resonance angiadhered to the offending vessels. Three patients showed recurrent ography and reconstructed imaging to confirm neurovascular symptoms. The isolation of Teflon pellets and offending vessels contact. In addition, patients underwent facial electromyography was performed carefully in 2 of these 3 patients; in 1 patient, such to assess whether they showed an abnormal muscle response. isolation could not be achieved. The intraoperative findings are 12 Surgeries were performed as previously described. In particular, summarized in Table 1. muscle dissections were performed carefully for patients who had not undergone cranioplasty. Surgical Outcomes All patients who underwent revision MVD surgeries were followed Evaluation After Revision Surgery up for 1 year. Nineteen patients experienced immediate relief of Patients’ postoperative symptoms were categorized into 2 groups: symptoms after secondary surgery. Twenty patients (86%) relief, no symptoms or intermittent symptoms without discomdemonstrated symptom relief at 1 year after secondary surgery. fort; and no relief, no improvement or symptom with discomfort. There was 1 complication, permanent facial palsy of HousePatients were reevaluated at 1 year postoperatively. Brackman grade III, in 1 patient. In most patients with Teflon malposition, symptoms were improved immediately postoperatively; all had excellent outcomes at 1 year after secondary RESULTS surgery. Among patients with incomplete decompression, 5 Patients’ Characteristics experienced symptomatic relief immediately postoperatively. In Of the 23 patients in this analysis, 8 were men and 15 were women. addition, after 1 year of follow-up, 7 patients had improved to Two of the 23 patients had a history of MVD surgery in our center; symptomatic relief. One patient with distal compression had no the remaining 21 patients had undergone MVD procedures in symptomatic relief at 1 year postoperatively. Finally, among paother hospitals. The mean age of the patients was 55.3 years tients with Teflon granuloma, improvements were noted in only 1 (range: 33e70 years), and the mean duration between initial MVD who had undergone removal of granuloma, at the time of 1-year surgery and revision surgery was 44.9 months. Sixteen patients follow-up visits (Table 2). Twenty-two patients showed abnormal had persistent symptoms after initial surgery, and 7 patients had muscle responses on preoperative facial electromyography; this
e2
www.SCIENCEDIRECT.com
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.07.191
ORIGINAL ARTICLE CHANG KYU PARK ET AL.
REVISION MVD FOR HEMIFACIAL SPASM
Table 1. Characteristics of Intraoperative Findings Intraoperative Finding
Table 2. Surgical Outcomes of Reoperation Immediately After Second MVD
Patients, n
1 Year After Second MVD
Teflon malposition Absent from REZ
5
CN IX
1
Intraoperative Finding
IAM
1
Teflon malposition
Vicinity of CN VII
5
Absent from REZ
5
5
0
5
0
Prior offender
2
CN IX
1
1
0
1
0
Perforators
3
IAM
1
1
0
1
0
2
Vicinity of CN VII
5
5
0
5
0
2
0
2
0
Incomplete decompression
Compression of distal portion New offender
1
Strong adhesion
3
n, number; REZ, root exit zone; CN, cranial nerve; IAM, internal acoustic meatus.
abnormal muscle response disappeared after the second MVD in 18 patients. Postoperatively, 19 of 20 patients with symptomatic relief showed no abnormal muscle response. However, 3 patients with persistent symptoms showed persistent abnormal muscle response. There was a statistical significance between the absence of symptoms and absence of abnormal muscle response (P ¼ 0.04). Relationship Among Preoperative Magnetic Resonance Imaging, Intraoperative Findings, and Surgical Outcomes On MRI of the Teflon malposition and incomplete decompression groups, neurovascular contacts that may cause symptoms were revealed near the REZ of the facial nerve (Figure 1). However, the patients with Teflon granulations did not have a clear neurovascular contact around the REZ (Figure 2). In most cases, prior Teflon was observed as a lesion with low signal intensity on T1-and T2-weighted images. Even in patients with granulation, no enhancement around the Teflon was observed. The presence of neurovascular contact in preoperative MRI was statistically significant for the surgical outcome (P ¼ 0.03), and the outcome of surgery was favorable when neurovascular contact was evident. DISCUSSION Recurrence and lack of symptomatic improvement comprise unwanted complications after MVD surgeries, such that the procedure appears ineffective. In addition, this increases confusion among patients who had previously undergone MVD surgery and experience recurrences. Although a few patients experience recurrences or show no symptomatic improvement despite surgery, the importance of MVD surgery for relieving HFS remains unquestionable and does not signify failure of the procedure. Goto et al6 initially described the delayed effect of the MVD procedure and suggested that follow-up should be performed for at least 1 year postoperatively to confirm the outcome.
WORLD NEUROSURGERY -: e1-e6, - 2019
Patients, Symptom Symptom n Relief Persistent Relief Persistent
Incomplete decompression Prior offender
2
Perforators
3
2
1
3
0
Compression of distal portion
2
0
2
1
1
New offender
1
1
0
1
0
Strong adhesion
3
1
2
1
2
n, number; MVD, microvascular decompression; REZ, root exit zone; CN, cranial nerve; IAM, internal acoustic meatus.
Similarly, Jo et al13 stated that the natural course after MVD surgery is sufficiently diverse that it is difficult to determine the optimal time for determining the failure or delayed effect of MVD surgery. Hence in this study, we included patients who underwent revision MVD surgeries for recurrences or lack of symptomatic improvement over 1 year of follow-up to assess delayed effects. In our study, we noted several potential sources of symptoms and observed significant differences between causes that led to persistent spasms and those that led to recurrent spasms. In patients who exhibited intraoperative malposition of Teflon pellets, persistent spasms after initial surgery could be attributed to incomplete or insufficient decompression of the offending vessels. Similarly, in patients with incomplete intraoperative findings, decompression had been performed using Teflon pellets; however, due to inadequate decompression, the offending vessels continuously excited the REZ, which caused persistent spasms postoperatively. In support of this theory, Yuan et al14 analyzed surgical outcomes in 23 patients who underwent revision MVD surgeries; they concluded that the incomplete decompression of the offending vessel was the main source of persistent symptoms after MVD surgery. Moreover, they categorized factors underlying incomplete decompression, which included inaccurate identification of offending vessels, inaccurate insertion of Teflon pellets, and insertion of inadequate volumes of decompressants. Similarly, Li et al15 described the findings in 1 patient with persistent symptoms after surgery: during revision MVD, other branches of the posterior inferior cerebellar artery
www.journals.elsevier.com/world-neurosurgery
e3
ORIGINAL ARTICLE CHANG KYU PARK ET AL.
REVISION MVD FOR HEMIFACIAL SPASM
Figure 1. Magnetic resonance imaging of the patient with Teflon malposition. (A) 3-Dimensional time-of-flight image showed neurovascular contact to the root entry zone of the right facial nerve (white arrow). (B)
were confirmed to compress the REZ. This may have been an underlying cause of persistent symptoms for patients in our study, although all underwent initial MVD procedures. In addition, all patients mentioned earlier showed recurrent symptoms after initial surgeries. The cause of recurrence was
Figure 2. Magnetic resonance imaging of the patient with Teflon granulation. (A) There was no neurovascular contact around the root entry zone (REZ) of left facial nerve (white arrow). (B) Prior Teflon with dark signal
e4
www.SCIENCEDIRECT.com
Prior Teflon with low signal density was revealed around IX and X cranial nerves (white asterisk). (C) T2-weighted image, Teflon was showed in same site (white asterisk).
either a new offending vessel or a new vascular contact point. Similarly, some patients initially showed disappearance of symptoms; however, these then recurred, combined with strong adhesion. Thus Li et al15 proposed that adhesion of Teflon pellets to the facial nerve acted in a manner similar to that of an offending
intensity was revealed in REZ of facial nerve (white arrowhead). A slight contrast enhancement was shown. (C) Dark signal intensity Teflon was shown in the same site.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.07.191
ORIGINAL ARTICLE CHANG KYU PARK ET AL.
vessel, thereby stimulating the nerve itself. Park et al9 suggested that the adhesion between decompressants and the nerve may contribute to recurrent HFS. In the present study, we were able to release the adhesion while avoiding nerve injury. As a result, patients who underwent removal of Teflon granulomas achieved favorite outcomes. Moreover, we confirmed that the Teflon granulation was the source of symptoms in patients who underwent MVD because we found facial nerve indentation after removal of Teflon granuloma. However, the circumstances under which granulation occurs remain unclear. There is no definite study about it. However, histologic studies of Teflon granulomas showed that their features were primarily associated with inflammation.11,16 Chen et al17 proposed a possible pathogenesis underlying the onset of Teflon granulomas: When Teflon contacts the dura or tentorium, a foreign body reaction is stimulated, comprising an initial inflammatory giant cell response. This leads to increased blood supply, followed by inflammatory reactions and development of granulation. In the present study, we analyzed the presence of Teflon in contact with the dura, as well as the presence of a vessel-rich environment around Teflon pellets in patients with Teflon granulomas; however, we found no definite association due to the small sample size. Therefore to prevent development of Teflon granulomas, it is recommended that small vessels be treated carefully to prevent bleeding and not to insert too much Teflon. The current study enrolled patients whose symptoms persisted for >1 year after initial surgery, and in whom we were able to attain effective results. However, if revision MVD is needed, there is no ideal timing for the procedure. Zhong et al18 reported success when revision MVD surgeries were performed early in patients with a history of failed initial MVD procedures; conversely, Wang et al19 obtained good results when performing delayed revision MVDs. Hence reviews of surgical outcomes by several authors suggest that the exact timing of revision MVD surgery is not particularly important; in contrast, the important aspect remains selection of patients who are most likely to benefit from the procedure. Park et al9 also performed revision MVD surgeries only in patients whose symptoms did not disappear after initial surgery, with offending vessels that could be confirmed in 3-dimensional time-of-flight magnetic resonance imaging. They reported considerable success following revision surgeries. Conversely, Yamaki et al20 performed revision MVD surgeries in 10 patients whose symptoms persisted or recurred after surgery; they reported that no key offending vessels could be identified intraoperatively and that surgical outcomes remained poor. In the present study, we performed revision MVDs in patients in whom clear or doubtful offending vessels could be identified in radiologic examinations; thus we were able to achieve good surgical outcomes to patients with clear neurovascular contact. Therefore decisions regarding revision MVD should be based on the presence of clinically specific symptoms, as well as on the evidence of offending vessels in radiologic examinations; this facilitates achievement of good surgical outcomes from the revision procedures. As mentioned previously, the identification of offending vessels in radiologic examinations forms the basis for revision surgery.
WORLD NEUROSURGERY -: e1-e6, - 2019
REVISION MVD FOR HEMIFACIAL SPASM
Despite this basis for surgery, intraoperative findings could be diverse. In the present study, we attempted to categorize intraoperative findings that were ultimately related to surgical outcomes. In patients in the first category (malpositioned Teflon pellets), no proper decompression of the REZ could be performed; hence if proper decompression can be performed, good surgical outcomes can be achieved. Similarly, in patients in the second category (incomplete decompression or compression from additional vessels), the cause of the spasm could be confirmed; if it was adequately and properly decompressed, good results could be achieved. However, when strong adhesion was observed between the nerve root and Teflon pellets, careful consideration was needed regarding continuation of the decompression procedure. In this situation, the adhesion itself can mimic the offending vessel, and relieving the adhesion can facilitate decompression and good results; however, extensive dissection to relieve the adhesion could also lead to injuries to the nerve, blood vessel, or both. These injuries can ultimately only result in unfavorable outcomes; thus proper visualization and understanding of the adhesion are essential prior to continuation of surgery. In the present study, strong adhesions were noted in 3 patients; no further surgical intervention was performed in 1 of these patients. Notably, no symptomatic recovery could be expected in this patient. In summary, good surgical outcomes can be achieved if adequate decompressive procedures are performed when Teflon pellets are not inserted at the REZ or in its general vicinity. These patients have a clear neurovascular contact on preoperative MRI. In patients with severe Teflon adhesion, there is no clear neurovascular contact around the REZ in the preoperative MRI, although severe adhesion or granulation may be present even if no enhancement is observed. In patients with these imaging characteristics, the success rate of revision MVD may be lowered. A considerable limitation of this study is that there were no previous operative records regarding initial surgeries performed at external hospitals; thus we were unable to confirm the types of decompressants used during initial surgeries, particularly in patients with intraoperative nonidentifiable decompressing agents. If proper identification could have been made of the types of decompressants used in the initial surgeries, intraoperative approaches might have been adjusted accordingly.
CONCLUSIONS In patients who have undergone MVD surgeries for HFS but persistent or recurrent symptoms remain, revision MVD procedures can be helpful if the symptoms can be clearly discerned both clinically and radiologically. Good surgical outcomes can be achieved if adequate decompression is performed in patients who have no Teflon pellets at the REZ or in its general vicinity, or if inadequate and improper decompressions were performed involving nonoffending blood vessels. And if a clear neurovascular contact is observed in the preoperative MRI, a favorable surgical outcome can be predicted.
www.journals.elsevier.com/world-neurosurgery
e5
ORIGINAL ARTICLE CHANG KYU PARK ET AL.
REFERENCES 1. Jannetta PJ, Abbasy M, Maroon JC, et al. Etiology and definitive microsurgical treatment of hemifacial spasm. Operative techniques and results in 47 patients. J Neurosurg. 1977;47:321-328. 2. Ma Q, Zhang W, Li G, et al. Analysis of therapeutic effect of microvascular decompression surgery on idiopathic hemifacial spasm. J Craniofac Surg. 2014;25:1810-1813. 3. Xia L, Zhong J, Zhu J, et al. Delayed relief of hemifacial spasm after microvascular decompression. J Craniofac Surg. 2015;26:408-410. 4. Zhong J, Zhu J, Sun H, et al. Microvascular decompression surgery: surgical principles and technical nuances based on 4000 cases. Neurol Res. 2014;36:882-893. 5. Barker FG 2nd, Jannetta PJ, Bissonette DJ, et al. Microvascular decompression for hemifacial spasm. J Neurosurg. 1995;82:201-210. 6. Goto Y, Matsushima T, Natori Y, et al. Delayed effects of the microvascular decompression on hemifacial spasm: a retrospective study of 131 consecutive operated cases. Neurol Res. 2002;24: 296-300. 7. Kureshi SA, Wilkins RH. Posterior fossa reexploration for persistent or recurrent trigeminal neuralgia or hemifacial spasm: surgical findings and therapeutic implications. Neurosurgery. 1998;43: 1111-1117. 8. Ravina K, Strickland BA, Rennert RC, et al. Revision microvascular decompression for
e6
www.SCIENCEDIRECT.com
REVISION MVD FOR HEMIFACIAL SPASM
trigeminal neuralgia and hemifacial spasm: factors associated with surgical failure. J Neurol Surg B Skull Base. 2019;80:31-39.
16. Deep NL, Graffeo CS, Copeland WR 3rd, et al. Teflon granulomas mimicking cerebellopontine angle tumors following microvascular decompression. Laryngoscope. 2017;127:715-719.
9. Park YS, Chang JH, Cho J, et al. Reoperation for persistent or recurrent hemifacial spasm after microvascular decompression. Neurosurgery. 2006; 58:1162-1167.
17. Chen J, Lee S, Lui T, et al. Teflon granuloma after microvascular decompression for trigeminal neuralgia. Surg Neurol. 2000;53:281-287.
10. Fan ZM, Han YC, Li JF, et al. [Revision surgery of the posterior fossa cranial nerves disease]. Zhonghua Er Bi Yan Hou Tou Jing Wai Ke Za Zhi. 2009; 44:364-368.
18. Zhong J, Zhu J, Li ST, et al. An analysis of failed microvascular decompression in patients with hemifacial spasm: focused on the early reoperative findings. Acta Neurochir (Wien). 2010;152:2119-2123.
11. Oda K, Higuchi T, Murai Y, et al. Teflon granuloma after microvascular decompression for hemifacial spasm: a case report and literature review. Neurosurg Rev. 2017;40:513-516. 12. Kim JP, Park BJ, Choi SK, et al. Microvascular decompression for hemifacial spasm associated with vertebrobasilar artery. J Kor Neurosurg Soc. 2008;44:131-135. 13. Jo KW, Kong DS, Park K. Microvascular decompression for hemifacial spasm: long-term outcome and prognostic factors, with emphasis on delayed cure. Neurosurg Rev. 2013;36:297-301.
19. Wang X, Thirumala PD, Shah A, et al. Microvascular decompression for hemifacial spasm: focus on late reoperation. Neurosurg Rev. 2013;36:637-643. 20. Yamaki T, Hashi K, Niwa J, et al. Results of reoperation for failed microvascular decompression. Acta Neurochir (Wien). 1992;115:1-7.
Conflict of interest statement: This work was supported by a grant from Kyung Hee University in 2018 (KHU-20182176). The authors declare that the article content was composed in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Received 15 May 2019; accepted 26 July 2019
14. Yuan Y, Zhang L, Zhang SX, et al. [Causes of ineffectiveness of microvascular decompression for hemifacial spasm]. Zhonghua Wai Ke Za Zhi. 2003;41:362-364.
Citation: World Neurosurg. (2019). https://doi.org/10.1016/j.wneu.2019.07.191
15. Li CS. Varied patterns of postoperative course of disappearance of hemifacial spasm after microvascular decompression. Acta Neurochir (Wien). 2005;147:617-620.
Available online: www.sciencedirect.com
Journal homepage: www.journals.elsevier.com/worldneurosurgery 1878-8750/$ - see front matter ª 2019 Elsevier Inc. All rights reserved.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.07.191