The balloon shape is a parameter with a very strong impact on outcome, and balloon volume should be adjusted to this parameter. Persistent elliptical balloon shapes should raise consideration of aborting the procedure. There were no differences in outcomes between 60 seconds and longer compression times. The number of previous operations did not correlate with pain relief, but seemed to increase the risk of complications. Patients with multiple sclerosis seemed to obtain similar benefit from the procedure as do patients with classic trigeminal neuralgia.

Only about 60-70% of well selected patients with neuropathic pain syndromes of peripheral origin enjoy sufficient pain relief with spinal cord stimulation (SCS). Since recent animal experiments have demonstrated that the GABA-B receptor is pivotal in the effect of SCS on certain neuropathic symptoms, the use of baclofen as an adjunct to stimulation emerged as an option in patients not responding satisfactorily to SCS. Forty-eight patients with neuropathic pain of peripheral origin responding poorly to SCS were enrolled in a study with intrathecal baclofen; in a few cases adenosine was also tried. Twenty patients reported significant pain reduction at bolus trials and were offered implantation of a drug pump. Seven patients subsequently had pumps implanted together with SCS and four had pumps alone. Three patients had only peroral baclofen therapy as an adjunct to SCS. The 14 patients continuing with baclofen therapy as an adjunct to SCS, or alone, were followed for an average of 35 months after pump implant. The group with SCS+pump n=5; 2 explanted) reported an average decrease of pain ratings from VAS 82 to 33. The group with i.t. baclofen only had a pain decrease from VAS 63 to 33, while the three patients with peroral baclofen+SCS had less benefit from drug therapy. Adjunctive drug therapy for patients with unsatisfactory pain relief by SCS may offer a possibility to enhance pain alleviation.

In a previously published pilot study, we addressed the possibility to increase the effectiveness of spinal cord stimulation (SCS) applied for neuropathic pain by using adjunct pharmacological therapy. This combined treatment approach was a direct spin-off from animal experiments aiming at the exploration of transmitter and receptor mechanisms involved in the pain relieving effect of SCS. Out of 48 patients with neuropathic pain of peripheral origin responding poorly to SCS, seven received pumps for intrathecal baclofen (GABA-B receptor agonist) delivery together with SCS, and four had pumps alone. In order to assess the long-term effect a follow-up has been performed, with an average, total treatment time of 67 months. At the follow-up the remaining nine patients still enjoy about the same pain relief as initially, but with a mean, further dose increase of about 30%. This study demonstrates that a deficient SCS effect in neuropathic pain may be considerably improved by intrathecal baclofen administration, and that this enhanced effect persists for a long-time. On-going and future animal studies may provide new and even more efficient pharmaceutical candidates for such combined therapy.

Both PRGR and PBC are effective techniques for the treatment of trigeminal neuralgia, with PRGR presenting some advantages in terms of milder and fewer complications and allowing lighter anesthesia without compromise of analgesia. For these reasons the authors consider PRGR as the first option for the treatment of trigeminal neuralgia in patients who are not suitable candidates or are not willing to undergo microvascular decompression, while PBC is reserved for patients in whom the effect of PRGR has proven to be short or difficult to repeat due to cisternal fibrosis.

Implantation of laminotomy electrodes can be performed conveniently with spinal anesthesia because it minimizes discomfort for the patient and enables the use of intraoperative test stimulation to guide the positioning of the electrode. In spite of the total motor block and anesthesia, paresthesiae representing the activation of the dorsal columns can be evoked and are well perceived, and the thresholds are not abnormally high. This observation supports the notion that the subarachnoidal anesthetic agent acts predominantly on the spinal rootlets rather than on the spinal afferent pathways.