Object. The goal of this study was to analyze prospectively factors associated with facial pain outcomes after percutaneous retrogasserian glycerol rhizotomy (PRGR) for patients with medically unresponsive idiopathic trigeminal neuralgia.
Methods. Between July 1999 and December 2003, 98 patients underwent PRGR in the manner described by Håkanson. The mean patient age was 72.1 years and the average pain duration prior to PRGR was 8.6 years. Fifty patients (51%) had previously undergone surgery. In six patients (6%), the trigeminal cistern could not be defined and the procedure was aborted; six patients were lost to follow-up review. An excellent facial pain outcome was defined as the absence of facial pain without medications.
Thirty-two (35%) of 92 patients either received no benefit (17 patients) or experienced recurrent pain (15 patients) and required additional surgery at a mean of 7.5 months after PRGR; the mean duration of follow-up review in the other 60 patients was 28.7 months (range 3–52 months). Including patients who did not receive a glycerol injection, the 1- and 3-year chances of an excellent facial pain outcome were 61 and 50%, respectively. A multivariate analysis of clinical and surgical factors showed that the facial pain exhibited on glycerol injection correlated with excellent facial pain outcomes (relative risk [RR] = 1.02; 95% confidence interval [CI] 0.26–1.77; p < 0.01), whereas patients who experienced any constant pain less frequently had excellent outcomes (RR = 1.13; 95% CI 0.06–2.20; p = 0.04). Forty-six patients (53%) experienced either mild numbness/parathesias (39 patients) or dysesthesias (seven patients). New trigeminal deficits after PRGR were associated with excellent facial pain outcomes (RR = 1.25; 95% CI 0.56–1.93; p < 0.001).
Conclusions. Percutaneous retrogasserian glycerol rhizotomy remains a good operation for patients with medically unresponsive trigeminal neuralgia who are considered poor candidates for posterior fossa exploration. Predictive factors for success include patients without any constant facial pain, patients with immediate facial pain during glycerol injection, and patients with new trigeminal deficits after PRGR.