Significance of Abnormal Muscle Response Monitoring During Microvascular Decompression for Hemifacial Spasm

Fig. 1

The AMR findings during surgery

Outcome Evaluation

The postoperative results were evaluated on the first day after the surgery and on the latest follow-up day. The follow-up period ranged from 11 to 62 months, with an average of 34.1 months. The postoperative result was regarded as “relief” when the spasms disappeared completely or the symptoms (frequency and degree of the attack) improved by more than 75 %; the postoperative result was regarded as “no relief” when the spasms decreased by less than 25 % or remained unchanged.

Statistical Analysis

Data processing was performed using commercially available software (SAS, version V8, North Carolina America). The Chi-square test and grouped t-test were used to assess the correlation between the intraoperative AMR and the clinical outcome. The level for statistical significance was a probability value of less than 0.05.

Results

AMR Findings

Among the 332 patients with a typical AMR wave recorded at the beginning of the operation, 6 had a negative AMR after durotomy and cerebrospinal fluid (CSF) drainage. With exposure of the REZ and transposition of the compressing artery, the AMR vanished in 176 patients. For the other 150 patients with remaining AMR, the entire VII cranial nerve root was dissected. After all the contacting vessels had been moved away, the AMR disappeared in 118 patients. Among the remaining 32 patients with positive AMR, the amplitude changed to <50 % of the baseline in 7, while it changed to ≥50 % in 25. For the latter group, neurocombing was performed. Finally, there were still 16 patients with AMR amplitude of ≥50 % of the baseline at the end of the operation (Fig. 1).

Correlation Between AMR and Postoperative Outcomes

In this series, AMR disappeared completely in 305 patients (91.9 %) and its amplitude decreased to less than 50 % of the baseline in 11 (3.3 %), while its amplitude remained at more than 50 % of the baseline in 16 (4.8 %). Of those with AMR < 50 %, 98.4 % achieved relief on the first postoperative day and the final follow-up day, while of those with AMR ≥ 50 %, relief was achieved in 18.8 % and 25 %, respectively, at these times (P < 0.01). Accordingly, a more than 50 % decrease of AMR amplitude may predict a good prognosis. The sensitivity, specificity, and accuracy of AMR monitoring during MVD were 99 %, 72.2 % and 97.5 %, respectively. The false-negative and false-positive rates were 0.96 % and 27.8 %, respectively (Table 1).

Table 1

Patient distribution in abnormal muscle response (AMR) ≥ or <50 % groups according to postoperative outcomes

Group	Postoperative		Follow-up		Total
Group	Relief	Non-relief	Relief	Non-relief	Total
AMR < 50 %	311	5	311	5	316
AMR ≥ 50 %	3	13	4	12	16
Total	314	18	315	17	332
P value	<0.05		<0.05		332

AMR < 50 %, patients with AMR disappearance plus those with AMR amplitude decreased by over 50 %; AMR ≥ 50 %, patients with AMR amplitude decreased by no more than 50 %

Complications

There was no mortality and no severe complication occurred postoperatively, with the exception of transient conductive-type hearing impairment in three patients (0.9 %), immediate facial weakness in three (0.9 %), and delayed facial palsy in six (1.8 %).

Discussion

Several investigators have attempted to determine whether AMR monitoring during MVD leads to positive outcomes. Some authors believed that intraoperative monitoring of the AMR was useful for identifying the offending vessels and for confirming a successful decompression of the facial nerve [3, 13]. It has been reported that the chance of cure in those with AMR disappearance was 4.2 times that in patients with AMR persistence [11]. An analysis of 1301 cases with a 3- to 6-month follow-up showed that complete disappearance of AMR was associated with a higher postoperative cure rate. The sensitivity, specificity, positive predictive value, and negative predictive value were calculated as 80 %, 39 %, 89 %, and 24 %, respectively [14]. Some authors have stated that AMR may be an unreliable predictor of long-term outcome, since they found that those in the non-AMR-disappeared group also had relief from HFS after MVD. For example, in a 90-patient series, AMR disappeared during surgery in 80, of whom 5 had persistent HFS, and 1 developed a recurrence of HFS. Of the 10 patients with persistent AMR despite effective MVD, 8 achieved complete resolution [4].

Nevertheless, we believe that AMR was a good tool to navigate the MVD process when it was used properly. Actually, this AMR is very sensitive, and could change as soon as the offending artery was separated from the facial nerve. That is the reason why the AMR disappeared in some cases even before the offending artery was visualized. Change of AMR may happen during the process of durotomy with CSF drainage or retraction of the cerebellum [10]. Evidently the AMR is not reliable until you find a compressing artery. As a matter of fact, the AMR is not used to search for the culprit but to confirm your finding. Sometimes the disappearance of AMR may be delayed for a couple of minutes, which could lead to a false-negative result. Therefore, in order to achieve a better result with fewer complications, our surgical strategy of MVD is to dissect the caudal REZ of the facial nerve root first, because, in this area, the chance of finding the offending vessel is more than 90 % according to the literature [22]. Accordingly, we suggest, if you find an apparent compressing artery with a dent in the nerve in REZ, you just move the offending artery away and put some Teflon between the artery and the nerve for separation. Afterwards, you check the electrophysiology; if the AMR wave is abolished you can get ready to finish the operation. In this way, you may obtain resolution of HFS in most cases with minimal risk (our data exhibited fewer complications compared with data in the literature [19, 22]). However, the monitoring should not be ceased at this stage. If the AMR reappears, a double check is necessary. In that case, one should not hesitate to reopen the incision, because the offending artery may have contacted with the nerve again while the retracted cerebellum was released. On the other hand, if an offending artery is not discovered in the REZ, then an entire dissection of the intracranial nerve root is strongly suggested. You should separate all the contacting vessels away from the nerve. After that, if the AMR remains, a combing of the facial nerve could be added. In consideration of a persistent AMR after a thorough decompression, we recommended a new criterion: a more than 50 % decrease of AMR amplitude (Fig. 1). Our study showed this criterion has high accuracy.

Only gold members can continue reading. Log In or Register to continue