BACKGROUND: Microvascular decompression is an accepted treatment for primary trigeminal neuralgia. In recent years, efforts have been made to investigate the role of preoperative MRI/MRA in selecting patients with neurovascular compression. METHODS: Ninety-two consecutive patients underwent a standard retrosigmoid craniotomy for microvascular decompression between 2005 and 2009, of whom 67 had a preoperative MRI and MRA according to the protocol of our department. Operative findings were accurately recorded and compared to MRI/MRA results. Follow-up was in the form of telephone interview. RESULTS: Preoperative MRI sensitivity was 96%, while specificity was 75% (1 false positive among 4 patients with negative intra-operative findings). The predictive value of a positive MRI/MRA was 98%, while the predictive value of a negative MRI/MRA was 50%. Fifty-four patients were available for outcome assessment at a mean follow-up of 3.8 years (range 1-5). Seventy-two percent of patients were pain-free, 9.5% had mild residual pain, while 18.5% had a poor outcome (moderate to severe residual pain). The correlation between preoperative MRI/MRA and outcome was not statistically significant (P = 0.570). CONCLUSIONS: Preoperative MRI has both good sensitivity and positive predictive value. Specificity and negative predicitve value were limited in this series. No correlation was found between preoperative imaging and outcome. Both radiological and clinical criteria are important for patient selection.
BACKGROUND: Microvascular decompression is an accepted treatment for primary trigeminal neuralgia. In recent years, efforts have been made to investigate the role of preoperative MRI/MRA in selecting patients with neurovascular compression. METHODS: Ninety-two consecutive patients underwent a standard retrosigmoid craniotomy for microvascular decompression between 2005 and 2009, of whom 67 had a preoperative MRI and MRA according to the protocol of our department. Operative findings were accurately recorded and compared to MRI/MRA results. Follow-up was in the form of telephone interview. RESULTS: Preoperative MRI sensitivity was 96%, while specificity was 75% (1 false positive among 4 patients with negative intra-operative findings). The predictive value of a positive MRI/MRA was 98%, while the predictive value of a negative MRI/MRA was 50%. Fifty-four patients were available for outcome assessment at a mean follow-up of 3.8 years (range 1-5). Seventy-two percent of patients were pain-free, 9.5% had mild residual pain, while 18.5% had a poor outcome (moderate to severe residual pain). The correlation between preoperative MRI/MRA and outcome was not statistically significant (P = 0.570). CONCLUSIONS: Preoperative MRI has both good sensitivity and positive predictive value. Specificity and negative predicitve value were limited in this series. No correlation was found between preoperative imaging and outcome. Both radiological and clinical criteria are important for patient selection.