Hemifacial spasm (HFS), a syndrome that involves unilateral involuntary twitching and contractions of facial muscles, is caused by vascular compression of the facial nerve at the root exit zone. Microvascular decompression (MVD) has been established as an effective surgical procedure for hemifacial spasm1,18). In patients with HFS, the lateral spread response (LSR), produced by electrical stimulation of one branch of the facial nerves, is recorded from facial muscles after facial nerve stimulation. M©ªller and Jannetta12-14) eloquently described the LSR together with the use of an intra-operative evoked electromyogram (EMG) during MVD procedures. Some authors reported that the disappearance or decreased amplitude of the LSR after MVD was associated with postoperative spasm relief or a favorable clinical outcome6,19). On the other hand, other authors suggested that the disappearance or persistence of an abnormal response in the intra-operative monitoring after decompression cannot predict the improvement of facial spasm and its effectiveness may be questionable3,7,9). There have been many reports on the association between LSR findings after decompression and the clinical outcome of hemifacial spasm.

Few reports described the association between the disappearance of LSR before and after MVD and the post-operative result of spasm. Kim et al.8) reported that when comparing two groups based upon whether the LSR disappeared before or after decompression, facial EMG monitoring of the LSR is helpful in predicting outcomes. This factor must be considered a prognostic factor of HFS after MVD, together with the disappearance or persistence of the LSR after decompression. In that paper, the mean follow-up duration was 17.9 months (range : 12-27 months) less. Reports on the correlation between the disappearance or persistence of the LSR after MVD and clinical outcomes usually had a follow-up duration of less than about two years5,7,17). In the authors' institute, their senior colleague (S.H.L.) has also checked continuous intra-operative LSR monitoring during MVD and the clinical outcomes of hemifacial spasm. Furthermore, long-term clinical follow-up data over two years were collected on the intra-operative monitoring results.

It was hypothesized that the disappearance of the LSR before and after MVD also predicts the long-term clinical outcomes over two years. This study was conducted to clarify the long-term effectiveness of intra-operative electromyography during MVD for HFS. In the authors' prospective study, we investigated the association between the disappearance of LSR before and after MVD for the prediction of the short- and long-term clinical outcomes of HFS.

LSR is usually immediately reduced with decompression of the facial nerve root. There is still much controversy, however, over whether or not intra-operative LSR disappearance means adequate decompression of the facial nerve. Some authors3,7,15) reported that an intra-operative change in the LSR did not always indicate a favorable prognosis. Many authors tried to investigate the correlation between the disappearance or persistence of the LSR after decompression and the clinical outcomes with a mean follow-up duration of about two years5,7,17). To authors' knowledge, two papers report that the disappearance or decreased amplitude of the LSR indicates post-operative spasm resolution with a follow-up duration of over two years6,9).

On the other hand, similar to the authors' hypotheses, some studies have tried to prove spasm-free results by disappearance of the LSR before or after the decompression2,14,15,19). It was reported that LSR disappeared before the decompression of the compressing vessel in two of eight patients in one study2). The outflow of cerebrospinal fluid shifts the neurovascular relation, which is temporarily equivalent to decompression2). Mooij et al.15) deemed the abnormal muscle response (AMR) as indirectly confirming the AMR disappearance in patients after the drainage of the cerebrospinal fluid. Their results showed five (6.8%) patients in which the AMR disappeared before the decompression. These five patients with indirect confirming had a lower rate of cured spasm than 25 (33.8%) patients with guiding in which decompression was followed by disappearance of the AMR. Yamashita et al.19) described the AMR disappearance in 53 of 60 patients after their microsurgery, and in nine patients before the transposition of the offending arteries. Among these nine patients, three (33.3%) showed persistent facial spasm in their immediate results. In the long-term results, however, the nine patients were completely cured. These three reports announced the LSR disappearance before the decompression, but they did not directly compare the LSR disappearance before and after the decompression with the spasm-free outcome, and performed the statistical analysis. Kim et al.8) reported that the 75.6% complete cure rate of Group B (in which the LSR disappeared before the decompression) was much lower than the 92.9% of Group A (in which the LSR disappeared after the decompression). Moreover, the spasm-free outcomes in the three-month and one-year results had statistically significant differences between the two groups (p value <0.05). In the discussion section, it was mentioned that further long-term follow-up evaluation may provide more information regarding the association between intra-operative LSR monitoring and post-operative results.

Fortunately, we had opportunities to analyze the results for the two groups with long-term follow-up periods of more than two years. As mentioned, our study also revealed that the complete relief rate of Group A was higher than that of Group B, not only after a one-year follow-up but also after two- and three-year follow-ups. The post-surgical two- and three-year results also significantly differed between the two groups. In this study, the patients in whom the LSR disappeared before the decompression showed poorer results than those in whom the LSR disappeared after the decompression during the long-term follow-up periods of over two years.

Unlike previous results8), the three-month spasm-free results did not differ significantly. It may take time before the motor nucleus hyper-excitability of the facial nerve decreases and its re-myelination process is completed. Several authors have proposed that in some patients, once the vascular compression is resolved, the motor nucleus hyperactivity starts to decline slowly and normalizes over a few months to a few years4-6).

Considering the lower spasm relief rate in Group B (in which the LSR disappeared before the decompression) than in Group A (in which the LSR disappeared after the decompression), it cannot be intra-operatively confirmed if the LSR disappearance in the patients in Group B was actually due to the decompression of the conflicting vessel, because the LSR disappeared before the decompression. Thus, more careful and adequate decompression between the offending vessel and the facial nerve root exit zone may be needed to improve the clinical outcomes of MVD.

Although the LSR disappearance before MVD had poorer outcomes, it is not based on scientific evidence and cannot be proven logically. Moreover, in these patients, despite the knowledge of the need for more careful and adequate decompression, it is not known how much or where the Teflon felt must be added. Furthermore, a study to determine the scientific cause of this phenomenon may be needed.

In this study, it was found that patients in whom the LSR disappeared after the decompression had better prognoses than those in whom the LSR disappeared before the decompression in the long-term follow-up periods of over two years. Thus, intra-operative facial EMG monitoring is helpful in predicting HFS prognosis and verifying the decompression of the facial nerve.

In patients in whom the LSR disappeared before the decompression, the facial nerve should be more carefully and adequately decompressed.