Different Roles of Microvascular Decompression in Hemifacial Spasm and Trigeminal Neuralgia

Abstract

Objectives  Microvascular decompression (MVD) involves the same procedure for both hemifacial spasm (HFS) and trigeminal neuralgia (TN), the resulting clinical courses are different. The aim of this study was to compare differences in MVD operations for HFS and TN and to determine the consequences of these differences on the clinical courses of the two diseases.

Methods  The medical records of patients who underwent an MVD operation between January 1998 and March 2013 were reviewed.

Results  A total of 2,263 patients were enrolled, 222 had TN and 2,040 had HFS. In the HFS group, the estimated cure rates at postoperative years 1, 2, and 3 were 92.0, 93.4, and 93.6%. In the TN group, the estimated cure rates at postoperative years 1, 2, and 3 were 88.4, 89.4, and 91.4%. Comparison of the initial treatment response between the HFS and TN groups did not reveal any statistically significant difference ( p  = 0.338). Among the 1,749 HFS patients initially cured by MVD, 57 were worsened. Among the 181 TN patients treated by MVD, 43 were worsened, with ≥ BNI III (Barrow Neurological Institute pain intensity score III) 12 worsened to a BNI score of IV. Comparing the recurrence rate between the HFS and TN groups, there was a statistically significant difference for cases with ≥ BNI III ( p  < 0.001), but not in cases with ≥ BNI IV ( p  = 0.498).

Conclusion  MVD is a promising treatment for HFS. In contrast, MVD-treated TN had a risk of recurrence. The MVD operation should be carefully considered when applied as a treatment for TN patients.

Introduction

Hemifacial spasm (HFS) and trigeminal neuralgia (TN) are the most common types of cranial nerve hyperactivity disorders caused by vascular compression. ¹ Although the possible etiologies of these disorders differ, the most important and correctable etiology is compression of the cranial nerve by the vascular structures of posterior circulation through demyelination of the root entry/exit zone of the nerve. ²

Microvascular decompression (MVD) is the only surgical approach that directly treats the proposed etiology of HFS and TN and offers a relatively low-risk opportunity to treat cranial nerve hyperactivity syndromes. Although the MVD procedure is the same for both HFS and TN patients, the resulting clinical courses are different.

The aim of this study was to compare the differences in the MVD operations used to treat HFS and TN and to determine the impact of these differences on the clinical courses of the two diseases.

Materials and Methods

Patients and Data Collection

We retrospectively reviewed the medical records of 2,263 patients who underwent retromastoid suboccipital craniectomy (RmSOC) with MVD between January 1998 and March 2013. All surgeries were performed by a single surgeon at our institute.

Of the 2,263 patients initially enrolled in this study, 222 had TN and 2,040 had HFS. Patients with symptomatic HFS or TN secondary to tumors or vascular anomalies were excluded from this study. To evaluate the cure and recurrence rates, patients with less than 30 days of follow-up were excluded. Ultimately, 1,870 patients were enrolled in the HFS group and 198 patients were enrolled in the TN group. The recurrence rate was only analyzed in patients who had been initially cured as a result of the operation. However, to evaluate complications after the operation, we did not exclude patients with short-term follow-up data, as most complications tend to develop within 2 weeks of MVD. All data relevant to MVD-related complications were collected.

Preoperative Evaluation and Postoperative Follow-Up

For the preoperative evaluation, the following series of tests on patients were conducted: magnetic resonance imaging (MRI), magnetic resonance angiography (MRA), temporal bone computed tomography (CT), facial nerve conduction study (NCS), electromyography (EMG), brainstem auditory evoked potential (BAEP), pure tone audiometry (PTA), and speech discrimination score (SDS). A noncontrast brain CT scan was performed 1 day after the MVD procedure. Three to 7 days after MVD, follow-up PTA, SDS, and physical examination were performed by the otolaryngologist. Patients visited the outpatient clinic 2 to 4 weeks after discharge for the physician to obtain clinical results. Follow-up visits were then scheduled on a case-by-case basis.

Surgical Procedures and Intraoperative Monitoring

RmSOC was performed as described by McLaughlin and colleagues. ³ Briefly, a craniectomy with an opening the size of 20 × 25 mm was performed. After opening the dura mater, the cerebellum was gently retracted to expose the trigeminal or facial nerve. Several Teflon balls and threads were inserted between the corresponding cranial nerve and the offending vessels. If a mastoid air cell was opened, the opened air cell was meticulously sealed with bone wax and a muscle patch was placed on the open surface. ⁴ During surgery, BAEP and facial EMG monitoring were performed from the time of general anesthesia administration until dural closure.

Data Collection and Chronological Analysis

All patients were followed postoperatively at the outpatient department at regular 2 to 3 months intervals. Evaluation and recording of postoperative outcomes were performed by an assisting nurse and surgeon on postoperative days 1 and 3, week 3, months 3, 6, and 12, and more frequently when needed.

In the HFS group, a questionnaire regarding patient satisfaction and disease status was administered. The questionnaire included the rating of spasms on a scale from 0 to 10, with 10 indicating the worst spasm ever experienced in terms of both frequency and intensity. Cure was defined as no residual spasm and was associated with a score of 0 on this scale. Postoperative courses were categorized into five different groups according to the similarity of temporal changes in the residual symptoms ⁵ and symptomatic changes during each follow-up interval were chronologically analyzed among the five different groups. The five improvement patterns were designated as group A (immediate recovery without relapse), group B (temporary relapse followed by cure), group C (slow but steady improvement that led to cure after one or more months), group D (recurrence with sustained symptoms), and group E (no improvement or improvement to some extent that did not lead to cure). Groups A, B, and C were considered the successful groups, while groups D and E were considered the unsuccessful groups. Symptomatic changes during each follow-up interval were also measured to determine recurrence.

For the TN group, the classification scheme devised by a group at the Barrow Neurological Institute (BNI) was used. Briefly, the BNI pain intensity score is designated as score I (no pain, no medication required), score II (occasional pain, no medication required), score III (some pain, adequately controlled with medications), score IV (some pain, not adequately controlled with medications), or score V (severe pain or no pain relief). Scores I and II were considered indicative of a successful treatment, with scores III or more being indicative of an unsuccessful treatment. During follow-up, a relapse of pain was considered disease recurrence. Patients with recurrence were divided into two groups: those with a BNI score greater than III and those with a BNI score greater than IV.

The study protocol was reviewed and approved by the Institutional Review Board and adhered to the recommendations for biomedical research involving human subjects in the Declaration of Helsinki (1975). The requirement for informed consent was waived as the study was based on existing clinical data.

Statistical Analysis

Cohort summary data are presented using median (range) and number (%) where appropriate. Categorical characteristics were compared using the Chi-square test or Fisher's exact test. Estimated cure and recurrence rates were compared using the log-rank test with Kaplan–Meier curves. p -Values less than 0.05 were considered statistically significant. Statistical analyses were performed using commercial software (SPSS, version 20; IBM, Armonk, NY, U.S.A.).

Results

Demographics and Clinical Data

Of the 2,263 patients initially enrolled in this study, 222 had TN and 2,040 had HFS. Follow-up data were available from 2 weeks to 132 months with a median of 18.35 months. The patient population consisted of 676 men and 1,586 women who ranged in age from 18 to 80 years (median age: 50 years).

Of the 2,040 HFS patients, 602 were males and 1,438 were females with a male-to-female ratio of 0.42:1 and their median age was 50 years (range: 18–77 years) Meanwhile, among the 222 patients with TN, 74 were males and 148 were females with a male-to-female ratio of 0.51:1 and their median age was 54 years (range: 25–80 years). The median follow-up period was 18 months (range: 0.5–121 months) in the HFS group and 25 months (range: 0.5–130 months) in the TN group.

When comparing the patients with HFS and TN, there was no statistical difference in sex distribution. The age of the patients was significantly higher in the TN group than in the HFS group ( p  < 0.001), and the TN group also had longer follow-up periods compared with the HFS group ( p  < 0.001). There was a similar number of patients in the HFS and TN groups whose right and left sides were affected (0.96:1); however, the TN group had a higher number of patients whose right sides were affected. In the HFS group, the offending vessels were mostly the anterior inferior cerebellar artery (AICA) and/or posterior inferior cerebellar artery (PICA; 98.5%), while in the TN group, the superior cerebellar artery (SCA) was the predominant offending vessel (70.0%; Table 1 ).

Table 1. Demographic and clinical data of TN and HFS patients who underwent microvascular decompression surgery.

	Hemifacial spasm ( n  = 2,040)	Trigeminal neuralgia ( n  = 222)	p -Value
Sex: male/female, no. (ratio)	602:1,438 (0.42:1)	74:148 (0.51:1)	0.239 a
Age at surgery, y; median (range)	50 (18–77)	54 (25–80)	< 0.001 b
Side of operation: right/left; no. (ratio)	998:1,042 (0.96:1)	150:73 (2.05:1)	< 0.001 a
Median follow-up period, mo; median (range)	18 (0.5–121)	25 (0.5–130)	0.001 b
Offending vessels, %			< 0.001 c
SCA		156 (70.0%)
AICA	1,111 (54.5%)	12 (5.4%)
PICA	552 (27.1%)	2 (0.9%)
VA	23 (1.1%)	4 (1.8%)
SCA + AICA		11 (4.9%)
AICA + PICA	114 (5.6%)
AICA + VA	153 (7.5%)
PICA + VA	56 (2.7%)
AICA + PICA + VA	26 (1.3%)
V	2 (0.1%)	33 (14.8%)
Not clear	3 (0.1%)	5 (2.2%)

Abbreviations: AICA, anterior inferior cerebellar artery; HFS, hemifacial spasm; PICA, posterior inferior cerebellar artery; SCA, superior cerebellar artery; TN, trigeminal neuralgia; VA, vertebral artery; V, vein.

^aChi-square test.

^b Two-sample t -test.

^cFisher's exact test.

Surgical Outcomes of MVD for HFS and TN

Initial Treatment Response

The overall cure rate was calculated using the Kaplan–Meier method. For the 1,870 patients in the HFS group with over 30 days of follow-up data, the estimated cure rates at postoperative years 1, 2, and 3 were 92.0, 93.4, and 93.6%, respectively. In the HFS group, 121 patients had no treatment response to the MVD operation. For the 198 patients in the TN group with over 30 days of follow-up data, the estimated cure rates at postoperative years 1, 2, and 3 were 88.4, 89.4, and 91.4%, respectively. In the TN group, 17 patients had no treatment response to the MVD operation. Comparison of the initial treatment response between the HFS and TN groups revealed no statistically significant difference ( p  = 0.338; Fig. 1 ).

Fig. 1.

Initial response and estimated cure rate of MVD for HFS and TN. MVD, microvascular decompression; HFS, hemifacial spasm; TN, trigeminal neuralgia.

Recurrence Rate

Among the 1,749 HFS patients who had been initially cured by MVD, 57 worsened over the follow-up period, reporting difficulties with spasms and in daily living. Nineteen patients received botulinum toxin injections for retreatment. Seventeen patients are well treated with injection therapy but two patients are in discomfort condition and are considering reoperation. Most of the other 38 patients are observing with wax and wane symptoms but two patients are considering reoperation with discomfort. The estimated recurrence rates at postoperative year 1, 2, and 3, were 1.5, 3.3, and 9.1%, respectively. Median progression-free survival (PFS) was not reached and 75% PFS was estimated at 11.9 years. Among the 181 TN patients who had initially been cured by MVD, 43 worsened over the follow-up period with BNI scores greater than III; 12 of these patients had worsened to a BNI score of IV. Eleven patients received gamma-knife radiosurgery to control for recurrent pain. Seven patients were able to maintain pain free condition without medication after gamma-knife radiosurgery. However, the other four patients needed the medication afterwards. One patient received ablation therapy with radiofrequency and maintained pain free condition without any medication. The 5-year symptom-free survival was 37.9% for patients with a BNI score greater than III and 12.9% for patients with a BNI score greater than IV. The median PFS was 6.1 years (95% CI: 4.3–7.8) for those reporting a BNI score greater than III. In contrast, patients with a BNI score greater than IV did not reach the median PFS. When the recurrence rates between the HFS and TN groups were compared, a statistically significant difference was found for cases with a BNI score of III or greater ( p  < 0.001) but not for cases with a BNI score of IV or greater ( p  = 0.498; Fig. 2 ). Compared with the offender vessels in recurrent cases, there is no significant difference compared with the total group ( Table 2 ) .

Fig. 2.

Recurrence and estimated recurrence rate of MVD for HFS and TN. MVD, microvascular decompression; HFS, hemifacial spasm; TN, trigeminal neuralgia.

Table 2. Characteristics and management of recurrent cases after microvascular decompression surgery.

( n )	Offender	Single	Multiple	Vein	Management		Clinical result
Hemifacial spasm (57)		29	9	0
					Botox injection	19	17 cases well controlled
							two cases considering reoperation
					observation	38	two cases considering reoperation
Trigeminal neuralgia (43)
BIN III (31)		27	4	7	medication	31	all well controlled
BIN IV (12)		10	2	4	GKS	11	seven cases controlled
							four cases need medication
					RF	1	well controlled

Abbreviations: BIN, Barrow Neurological Institute; GKS, Gamma knife radiosurgery; RF, radiofrequency ablation.

Complications of MVD for HFS and TN

In the HFS group, delayed-onset facial nerve palsy (dFNP) was the most common complication (149 patients, 7.30%), followed by middle ear effusion (100 patients, 4.90%) and hearing loss (74 patients, 3.63%). In the TN group, the common complications were middle ear effusion (12 patients, 5.38%), hearing loss (three patients, 1.35%), dysgeusia (three patients, 1.35%), and wound revision surgery (three patients, 1.35%). When comparing the two groups, facial nerve palsy was the only complication that was statistically more common in the HFS group ( p  < 0.001). Profiles of the complications after MVD and the incidence of each complication are summarized in Table 3 and Fig. 3 .

Table 3. Postoperative complications in patients who underwent microvascular decompression surgery.

	Hemifacial spasm ( n  = 2,040)	Trigeminal neuralgia ( n  = 222)	p -Value
Facial nerve palsy			< 0.001 a
Immediate onset	18 (0.88%)	0
Delayed onset	149 (7.30%)	2 (0.90%)
Hearing loss (serviceable, nonserviceable)			0.074 a
Serviceable hearing loss	50 (2.45%)	2 (0.90%)
Nonserviceable hearing loss	24 (1.18%)	1 (0.45%)
Dysgeusia	8 (0.39%)	3 (1.35%)	0.085 b
Diplopia, 6th nerve palsy	2 (0.10%)	0	NA
Lower cranial nerve palsy	9 (0.44%)	0	NA
Symptoms related to CSF leakage	107 (5.25%)	10 (4.48%)	0.528 a
Middle ear effusion	100 (4.90%)	12 (5.38%)	0.743 a
CSF rhinorrhea	12 (0.59%)	0	NA
Meningitis	8 (0.39%)	1 (0.45%)	0.606 b
Revision surgery for wound problems	7 (0.34%)	3 (1.35%)	0.067 b
Vascular complications	5 (0.25%)	1 (0.45%)	0.462 b

Abbreviation: CSF, cerebrospinal fluid; NA, not available.

^aChi-square test.

^bFisher's exact test.

Fig. 3.

Differences in complications between HFS and TN patients. HFS, hemifacial spasm; TN, trigeminal neuralgia; CSF, cerebrospinal fluid.

Discussion

Differences in Clinical Course between HFS and TN Patients

MVD is the only way to provide a complete cure for HFS and TN, in contrast to toxin treatments that offer only temporary symptom relief. In particular, MVD for HFS aims to resolve compression of the facial nerve at the root entry or exit zone by aberrant or ectatic vessels. Previously, the success rates of MVD have been reported to range from 86 to 94%. ^{3 6 7 8 9} However, symptom recurrence has been reported to range from 2.4 to 10.3%. ^{9 10 11} Payner and Tew, Jr. ⁹ reported that 2 years after an MVD operation, the probability of recurrence is only 1.0%. In the present study, recurrence was found in 21 cases (1.2%) after 24 months of operation and four cases (0.2%) after 5 years of operation. Thus, recurrence appears to be quite rare after 24 months.

MVD has become one of the most common treatments for TN as it is capable of providing long-term pain relief. Unfortunately, not all patients have positive outcomes after MVD. Previous studies have shown that immediate postoperative pain relief occurs in 84 to 100% of patients. ^{12 13 14 15 16 17 18} The reported pain-free duration after MVD without medication ranges from 0.6 to 10 years. ¹⁹ After 1 year, the percentage of pain-free patients ranges from 80 to 91%; after 5 years, this percentage range drops to 58 to 72%. ^{13 14 15 17 18 20 21} In the present study, we observed recurrence 5 years after the initial operation in only one case with a BNI score greater than IV and in five cases with a BNI score greater than III. Recurrence, 5 years after the initial operation, is rare and most cases are well controlled by common medications.

Recurrence of HFS and TN after MVD is very challenging for neurosurgeons. According to previous studies, recurrence is more common when offending vessels are difficult to transpose or when decompression is obtained by inserting prostheses between the offending vessels and the facial nerve. ²² While recurrence may be associated with adhesions to or granulation of the prosthesis or to other issues, the causes of recurrence have not been clearly identified and should be elucidated in further studies.

Comparisons between patients undergoing MVD for either HFS or TN did not reveal any statistically significant differences in terms of initial treatment response. As our study was based on data from a single experienced surgeon, we expect that the applied surgical technique was well standardized. However, we observed a significant difference in recurrence rate between the two groups which suggests that differences in etiology may contribute to recurrence. Both HFS and TN are hyperactivity disorders of the cranial nerves. When these conditions are caused by vascular compression of the facial or trigeminal nerve roots, they are clinically regarded as idiopathic or primary HFS or TN. ^{7 18} While it is generally believed that HFS is caused by neurovascular compression, ⁶ the pathophysiology of TN remains elusive. Its etiology may even reflect a heterogeneous combination of several causes. Therefore, even though MVD is the only current treatment for TN, it may not address all etiologies of the disease.

The ratio of TN to HFS obtained in the current study contradicts that obtained in studies by western countries which may indicate a variable racial or geographic prevalence of the two diseases. To confirm the difference in the ratio, we reviewed medical records from the outpatient department between 2015 and 2016. Of the 172 patients visiting the institute with TN, 14 (8.1%) received an MVD operation, while 152 (36.3%) of the 419 patients with HFS received an MVD operation. A previous study showed that the percentage of Asians with HFS was higher than that of Hispanics, African-Americans, or Caucasians. ²³ While the reason for this difference is not clear, a possible explanation is the relatively small posterior fossa in Asian patients which makes vascular compression of the facial nerve more likely. ^{24 25} The higher prevalence of HFS may affect the difference in ratio, as well as the fact that TN can be more easily controlled with medication.

Treatment of TN remains a challenge for both neurologists and neurosurgeons. The first-line treatment for patients with TN is medical therapy with carbamazepine. The lack of clarity regarding the complex pathogenesis of TN remains a key factor in the unsatisfying results of medical therapy.

Differences in Complications between HFS and TN Patients

Anatomically, the cerebellopontine angle (CPA) is defined ventrally by the petrous part of temporal bone, cranially by the pons and middle cerebellar peduncle, caudally by the lobulus biventer and olive, and dorsally by the cerebellum. ²⁶ From a surgical point of view, the CPA consists of the cerebellopontine and cerebellomedullary cisterns. The etiology of cranial nerve hyperactivity syndrome (TN, HFS, paroxysmal positional vertigo, geniculate neuralgia, and glossopharyngeal neuralgia) has yet to be elucidated. Vascular compression of each nerve in the brainstem has been proposed to play a major role in the cause of cranial nerve hyperactivity through the demyelination of the root entry or exit zone of the nerve (i.e., ephaptic transmission). ² Unsurprisingly, postoperative complications for the same procedure in patients with TN and HFS may vary due to differences in the surgical corridors. In this study we observed that patients who received MVD for HFS, had a higher risk of facial nerve palsy and hearing loss than those who received MVD for TN. Details pertaining to complications have been reported in previous studies. ^{27 28 29 30 31}

Conclusion

The patients in the TN group were older than those in the HFS group. In the HFS group, both sides of the face of the patients were affected, whereas the right side of the face was affected more often in the TN group. The offending vessels were mainly the AICA and/or PICA in the HFS group and the SCA in the TN group. Comparison of initial treatment response between the HFS and TN groups did not reveal any statistically significant difference. Meanwhile, when comparing the recurrence rate between the HFS and TN groups, there was a statistically significant difference for cases with a BNI score of III or greater but not for cases with a BNI score of IV or greater. In the HFS group, delayed onset of facial nerve palsy was the most common complication (149 patients, 7.30%). In the TN group, the common complications were middle ear effusion (12 patients, 5.38%), hearing loss (three patients, 1.35%), dysgeusia (three patients, 1.35%), and wound revision surgery (three patients, 1.35%).

MVD is an effective treatment for both HFS and TN. Based on the findings of the present study, MVD is a very promising intervention for HFS but is associated with a risk of recurrence when used to treat TN. The application of MVD surgery should be carefully considered in the context of these specific conditions.