The long-term clinical outcomes of microvascular decompression for treatment of trigeminal neuralgia compressed by the vertebra-basilar artery: a case series review

Though most patients with trigeminal neuralgia (TN) do not have vertebrobasilar ectasia (VBE), a small percentage of TN cases may be associated with VBE. Most of these patients with VBE are characterized by significant dilation, elongation, and tortuosity of the vertebra-basilar arteries [1]. Pain results from the compression between the enlarged vessels and the trigeminal nerve at the base of the brain. Although a small dose of antiepileptic/antineuralgic drugs can initially provide excellent pain relief, up to 10–30% of patients do not respond to these drugs and need to undergo surgical treatment [2, 3]. The common surgical treatments for TN include ablative procedures, such as stereotactic radiosurgery and percutaneous rhizotomy, and non-ablative surgical microvascular decompression (MVD). MVD causes little or no facial numbness compared to percutaneous stereotactic rhizotomy and is highly successful as a gold standard first line treatment for TN with a minimal risk of pain recurrence [4, 5].

While MVD neurosurgery has been used to effectively treat TN due to the conflict contact between the trigeminal nerve and the vertebra-basilar artery (VBA), the long-term pain relief and the complications of the surgery need to be determined. The incidence and severity of MVD-associated complications seem largely dependent on the surgeon’s skill and knowledge of local surgical anatomy as well as the size and shape of the implants. In this study, we analyzed the long-term results of MVD treatment in 23 patients whose pre-operative pain scores were IV to V because of VBA compression, and we provided technical tips to minimize pain recurrence and potential complications of the neurosurgery.

MVD to separate the offending vessels from the affected trigeminal nerve is the most effective and accepted treatment for TN, and the technique has been reported to have a good success rate. Tatli et al. compared long term effectiveness of all surgical procedures for TN and demonstrated that success rate of immediate pain relief of MVD ranged from 76.4 to 98.2%, with 18.3 to 3% recurrence rate [9, 10]. Despite providing satisfactory results, MVD is not free of complications. Nineteen percent of the patients were reported as having hearing loss, and the mortality rate was 0.37% [10]. Other drawbacks of the MVD includes the recurrence rate, which has been reported to range from 3 to 30%, and implant-induced granuloma, which occurred in 1.3% of patients [10‐12]. The great variation of the recurrence and complication rates strongly suggests that the overall outcomes of treatment of TN with MVD surgery might be dependent on the surgeon’s skill and techniques, knowledge of local surgical anatomy, and the types of the wrapping materials in addition to the diagnosis, pre-operation preparation, and disease condition.

In this study, the percentage of the elder patients with TN was greater than in other studies [13]. Consistent with other reports, we also identified that the second and the third divisions of the left trigeminal nerve were more likely involved [14]. While the demographic and clinical characteristics of our patients and our immediate pain-free success rate after surgery were comparable with some reports [4, 14], the percentage of patients who had pain recurrences was lower, and cases that had permanent facial hypesthesia and developed a gradual hearing loss were fewer. We had no patients develop an implant related granuloma. Part of the difference in outcomes found could be explained by the fact that we had more typical patients (87%) as compared with other reports, and patients with a typical pattern of symptoms relapsed significantly less often and were less likely to develop severe complications after surgery [15]. We believe that minimized dissection, macrovascular mobilization and implant insertion between the offending VBA and the nerves in our surgery also contributed to the lower rate of side-effects.

We should point out that our sample size was small, and surgical procedure was in fact at low risk because of the patient selection criteria with relative low threshold. We noticed more complications in our work as the rates of transient complications were a little higher than those for standard MVD. This could be explained by our surgical procedures in which the arachnoid in cerebellopontine angle (CPA) needed to be open thoroughly to expose and mobilize the VBA, including the arachnoid covering cranial nerve VII-XII that should be preserve perfectly to protect these cranial nerves in standard MVD. In addition, MVD operation was done between these nerves, thus retraction was unavoidable. It is possible that these procedures could account for the higher transient postoperative complications. In addition, the postoperative pain relief seemed unusually delayed in this study. However, based on our experience delayed pain relief is commonly seen after MVD. The mechanism is not understood yet. It is generally believed that the treatment course and initial pain types (e.g. typical vs. atypical) may attribute to the delayed pain relief.

In this study, we used the 3D-TOF sequence to visualize the elongated, ectatic VBA and the twisted trigeminal nerve, which helped us determine the location and size of the incision for dissection and reposition of the large vessels in the MVD surgery and observe the position of Teflon implants and the surrounding tissue We prefer to place Teflon felt piece by piece between the VBA and the brain stem, gradually elevating the artery closer to the petrous bone and away from the REZ to decompress the nerve completely, and this approach achieved excellent results. However, we noted that a large volume of Teflon might be inserted into the posterior fossa to push the VBA far away from the trigeminal nerve, which increased the possibility of inflammation, adhesion, and TN recurrence after MVD. As evidenced in this study, we removed the extra Teflon and adhesion in one case and found that the patient’s facial pain immediately disappeared after the second surgery. For two cases whose VBAs were too long and hard, we developed a new strategy. We inserted Teflon between the vessel and the nerve so that the nerve was not completely decompressed as opposed to elevating the vessel away from the brain stem and decompressing the REZ for fear of touching the trigeminal nerve. We found that the insertion procedure was effective for those cases. We believe that distortion of the nerve would not cause TN because the nerve was still distorted after MVD but became relaxed. However, future studies are needed to assess the position and volume of Teflon implants as well as other methods that occupy less space to keep the nerve free and avoid recurrences. In addition, a study with a larger population and a longer follow up time is needed to confirm the findings.

While our success rate of immediate pain relief after surgery was comparable with some reports, the percentage of patients who had pain recurrences was lower, and cases that had permanent facial hypesthesia or developed a gradual hearing loss or other complications were fewer after MVD surgery. Our study indicates that MVD is an effective, reliable, and safe neurosurgery for treatment of TN compressed by the VBA. Failure of treatment and recurrence of the disease as well as complications could be minimized by preventing both the displacement of the Teflon implant and extraneous Teflon from touching the trigeminal nerves.