There have been many reports of percutaneous radiofrequency facet rhizotomy, perhaps better referred to as facet denervation, usually performed under general anaesthesia, with inconsistent success rates.

To report the authors' outcome data using both general and local anaesthesia and to reassess the value of this controversial procedure.

Our experience with 118 consecutive percutaneous radiofrequency facet rhizotomies performed on 90 patients in the Toronto Western Hospital was analyzed. Sixty percent of the procedures were performed under general anaesthesia, 40% under local anaesthesia. All patients had been temporarily virtually relieved of pain after local anaesthetic blockade of the subject facets by an independent radiologist.

The patients were monitored from 1 - 33 (mean 5.6) months after surgery, with complete elimination or a greater than 50% subjective reduction of pain considered the criteria for success. For the first or only procedure this was 41% overall, 37% in cases done under local anaesthesia, 46% in cases done under general anaesthesia (difference not statistically significant p=0.52). There was no statistically significant difference in success rates for procedures performed in the cervical, thoracic or lumbosacral facets, with unilateral versus bilateral denervations, when two to three as compared with more than three facets were denervated, nor for operations done in patients who had had previous spinal surgery compared with those who had not. Results were not better regardless of whether hyperextension of the spine aggravated the patient's preoperative pain or not, and when the procedures were repeated in the same patient outcomes tended to be consistent, arguing against repetition of failed facet denervations. The morbidity was low, the chief problem being sensory loss and transient neuropathic pain in the distribution of cutaneous branches of posterior rami in the cervical and thoracic areas; mortality was zero.

Percutaneous radiofrequency facet denervation is simple and safe, still worth considering in patients with disabling spinal pain that fails to respond to conservative treatment. The use of general anaesthesia shortens the operating time and the patient's discomfort without impairing success rate.

CT and MR guided stereotactic techniques have provided promising results in the management of brain abscesses. We reviewed our results of stereotactic management of brain abscesses in 20 consecutive patients with 28 abscesses from 1986 to 1993.

13 abscesses were in the cerebral hemispheres, 12 in the cerebellum, 2 in the pons and 1 in the thalamus. The bacterial organism was isolated in 12 of the 20 cases. All patients, except one who had a tuberculous abscess, were on antibiotics for less than 7 weeks.

Although there were 3 patients in coma before surgery, the mortality rate was zero and 17 patients had an excellent recovery with 3 patients having a persistent mild neurologic disability. Stereotactic aspiration of the largest lesion in the patients with multiple brain abscesses combined with intravenous antibiotic therapy was sufficient for the resolution of all lesions. Two of our patients treated with antibiotics alone showed abscess progression with neurologic worsening.

Stereotactic aspiration is safe, accurate, and when combined with the appropriate antibiotics, should be considered the procedure of choice in the management of brain abscesses.