Usefulness of intraoperative monitoring in microvascular decompression for hemifacial spasm: a systematic review and meta-analysis.

To review the diagnostic accuracy and possible added value of Brainstem Auditory Evoked Potentials (BAEP) monitoring and Lateral Spread Response (LSR) monitoring in microvascular decompression surgery for hemifacial spasms. For this systematic review we followed the PRISMA guidelines. We searched different databases and bibliographies of articles. We included studies on BAEP and LSR monitoring that reported data on hearing outcome or efficacy. Selected studies were assessed for bias using the MINORS tool. 64 articles were selected for qualitative synthesis, 42 met inclusion criteria for meta-analysis. The overall incidence of hearing loss was 3.4%. For BAEP monitoring AUC and pooled OR with 95% confidence interval were 0.911 (0.753–0.933) and 7.99 (3.85–16.60) respectively. Short-term data on LSR monitoring showed an overall spasm relief rate of 89% with pooled OR, sensitivity and specificity with a 95% confidence interval of 8.80 (4.82–16.08), 0.911 (0.863–0.943) and 0.451 (0.342–0.564) respectively. Long-term data on LSR monitoring showed an overall spasm relief rate of 95% with pooled OR, sensitivity and specificity with a 95% confidence interval of 4.06 (2.15–7.64), 0.871 (0.817–0.911) and 0.39 (0.294–0.495) respectively. The alarm criteria, a wave V latency prolongation of 1ms or a wave V amplitude decrement of 50%, proposed by the 'American Clinical Neurophysiology Society' are a sensitive predictor for postoperative hearing loss. Other BAEP wave changes, for example, complete loss of wave V, are more specific but correspond to irreversible damage and are therefore not useful as warning criteria. LSR monitoring has high diagnostic accuracy at short-term follow-up. At long-term follow-up, diagnostic accuracy decreases because most patients get spasm relief regardless of their LSR status. LSR persistence after surgery has a good long-term outcome, as long as an extensive exploration of the facial nerve has been performed. [ABSTRACT FROM AUTHOR]