Etiopathogenetic Considerations

A peripheral hypothesis (40, 64), a central hypothesis (19), and, more recently, theories supporting central-peripheral hypotheses (27, 60) for TN etiopathogenesis have been proposed. Nevertheless, it remains a puzzling mystery. Both trigeminal nerve lesions and central lesions affecting trigeminal pathways (MS and ischemia) (4, 87) have been reported to play an etiopathogenetic role in TN. Vascular cross compression is now increasingly accepted as an important etiological factor. We found a vascular conflict in most cases, even in patients with MS. Sometimes the involved vessels are subtle, and the root does not seem grossly compressed. Our MR data definitively demonstrate that the involvement of trigeminal pathways within the brainstem is very common in TN-MS patients. It is possible that demyelination of trigeminal fibers at the level of trigeminal root entry zone in the case of vascular cross compression (27, 34, 37, 40, 51, 64) and demyelination of the trigeminal pathways within the brainstem in the case of MS (59) may result in abnormal ephaptic transmission of impulses.

We found that vascular conflict (and possible consequent demyelination) and MS demyelination can coexist and that they may cooperate in the genesis of painful attacks. The classic distinction between the supposed “all central” mechanism for MS-associated TN and the “all peripheral” mechanism for the vascular compression-related TN should, thus, come under reconsideration. In its place we offer a unique (TN-MS patients are included), mixed central-peripheral mechanism in which abnormal impulses arise from demyelinated axons (MS, vascular compression, and any other possible cause of demyelination along the central and the peripheral course of trigeminal axons) and modulate the nuclear activity. Minimum myelin damage, without any nerve hypofunction, might be involved in the etiopathogenesis of idiopathic TN (19). Major myelin damage may be responsible for MS-associated TN based on the finding of possible clinical signs of trigeminal nerve hypofunction (87), magnetic resonance imaging (MRI) signs of demyelination, and, unfortunately, by the recurrence of pain after MVD. The concept of a central neuromodulatory role of impulses coming from the area of cross compression also explains the possibility that a long lasting alteration of discharge modalities of the trigeminal root can cause a decrease in the pain threshold, as suggested by recent reports on extracranial neurovascular conflicts (24, 67).

If this mixed peripheral-central hypothesis seems to be compatible with our (9) and others (2) apparently contradictory findings in TN, an alternative all-central hypothesis might also be considered. Supporters of this all-central mechanism deny any pathological role for vascular compression. According to this view, MVD elicits pain relief because it produces a sufficient trauma that interferes with normal nerve functioning, which then dampens the abnormal brainstem activity responsible for TN (17). In our series, we were not able to identify any prognostic factor. In particular, no statistically significant difference in the outcome between patients with severe versus mild conflicts was found that we believe adds further emphasis to the major role played by central mechanisms in patients with MS-related TN.

However, MVD certainly interferes with the pathological impulses that arise from the region of demyelination induced by chronic vascular cross compression, and even if it could not be considered as the definitive etiological cure (83), it is the only therapeutic option able to obtain pain relief without causing any sensory disturbance.

Percutaneous Methods

Since Hartel introduced his simple and direct percutaneous approach to the foramen ovale and gasserian ganglion in 1911 (32), several different methods to create therapeutic damage to the trigeminal root and ganglion became available. To reduce trigeminal sensory input, chemical agents such as alcohol, phenol, and glycerol with or without phenol were used. Possible diffusion of more aggressive neurolytic agents, such as alcohol, to untargeted structures and different individual responses to chemical neurolysis made the results of the injection of chemicals into the trigeminal cistern and ganglion quite unpredictable. Because an unfavorable recurrence rate and a frequent incidence of side effects, these techniques were progressively abandoned in favor of controlled radiofrequency thermal rhizotomy and mechanical balloon microcompression.

Radiofrequency Retrogasserian Controlled Thermorhizotomy

Radiofrequency retrogasserian controlled thermorhizotomy (TRZ) became the widely preferred treatment for TN after Sweet and Wepsic(79) introduced this technique in 1974. In the following years, experimental data supporting the effectiveness of TRZ for the differential destruction of small diameter nerve fibers have been reported (10, 18, 25), and its efficacy has been confirmed by many authors in large series of patients (3, 11, 23, 55, 70–72, 85). These experimental and clinical data showed that TRZ allows for sparing of the majority of facial touch sensibility, and hypalgesia or analgesia generally involves only the targeted trigeminal branches. More than 1700 patients have been treated at Instituto Nazionale Neurologico Carlo Besta since 1974. We were able to follow up with 97% of patients for a time ranging from 2 to15 years (mean follow-up, 72 mo); 71% of patients were found to be completely pain free without medication, 11% pain free with a small dose of antineuralgic drugs, and 15% still experiencing severe pain requiring a large dose of drugs or surgery (Table 37.3).

Regarding the amount of the inflicted sensory deficit, our data suggest that induced postoperative analgesia prevents the recurrence of pain in most of patients. In other words, patients with postoperative hypalgesia have a pain recurrence probability of 41% versus 7.5% for patients with postoperative analgesia. In all patients, the sensory deficit tends to diminish with time; nevertheless, a high percentage of patients with the more severe sensory postoperative deficit (analgesic patients) complain of dysesthesias. The total percentage of patients who required drugs for severe dysesthesia was 5%, with 1.5% of painful anesthesia that we were never able to definitively alleviate by any of the more advanced surgical antalgic techniques (open or percutaneous trigeminal tractotomy, trigeminal stimulation, cortical stimulation, deep brain stimulation, or CSF direct drug infusion). These complications are clearly related to the technique itself and cannot be completed avoided, even with meticulous surgical techniques, especially in the cases requiring repeated TRZ (Fig. 37.5).

However, by monitoring the corneal reflex during the procedure, major ocular deafferentation complications can be generally avoided, and keratitis requiring tarsorrhaphy was observed in only 0.5% of patients, even when the involvement of first branch was not considered as a contraindication to this kind of surgery. Masseter weakness with minor chewing impairment appeared in 10% of patients, whereas ocular palsy and diplopia appeared in 0.5%. Major neurological morbidity caused by intracranial bleeding was never observed. Mortality was null. This method can be proposed to patients accepting the risk of sensory disturbances when previous less-aggressive procedures have failed.

Balloon Microcompression of the Gasserian Ganglion

The observation that deliberate direct compression of the trigeminal ganglion was able to relief trigeminal pain by Sheldon in 1955 (69) led Mullan, in 1978, to develop a percutaneous relief controlled compression of the trigeminal ganglion that could be performed under short general anesthetic (56). The result that we were able to obtain by using balloon microcompression of the gasserian ganglion (percutaneous microcompression [PMC]) in 235 patients operated on since 1992 are reported in Table 37.4. The end point for compression was the achievement of a pear-shaped balloon in the cavum Meckel (Fig. 37.6). The balloon was then maintained inflated for approximately 1 minute. A longer compression resulted in a profound hypoesthesia that often led to the complaining of dysesthesias. Results derived from the literature are summarized in Table 37.5. This method seems to have the same limitations that characterize trigeminal surgery whatever the lesional procedure used, that is, the more the sensorial deficit, the longer the pain-free interval but the more frequent the severe dysesthesia.

However, PMC is easy to perform, and the recurrence rate is acceptable with a low rate of complication, even in the case of repeated surgery. Diplopia was sometimes observed, but it was generally transient. Because painful anesthesia and keratitis seem, in our opinion, too high of a price to be paid for pain relief, this is now the method we prefer when MVD fails or is refused by the patient.

Radiosurgery

Stereotactic gamma-knife radiosurgery (Electra, Sweden) was first reported for the treatment of trigeminal neuralgia by Leksell in 1971 (47). Its use, however, remained restricted to few centers until the mid-1990s when it started to become more widely used. Radiosurgical treatment of TN has been well investigated with gamma-knife devices involving fixed cobalt sources. Few reports exist concerning TN treated using linear accelerator (LINAC)-based devices. In recent years, these devices have reached the level of mechanical precision that is required for such functional treatments. Only one study reporting on patients treated with CyberKnife (Accuray, Sunnyvale, CA) is available in the literature (65). In our Institution, CyberKnife has been available since March 2004 (Fig. 37.7), but our data are too preliminary to be reported. Substantial advantages have been supposed in safety and comfort over other modalities, but the evidence is based on case series with a single randomized study comparing two methods of delivery of radiotherapy (63). The results obtained in some of the more significant series of the literature are reported in Table 37.6. From the analysis of the literature, the following conclusions can be drawn:

• Radiosurgery should be considered as a lesional procedure.
• A strong correlation between the development of new facial sensory loss and achievement and maintenance of pain relief after this procedure has been described (63).
• Quality of data is generally poor: case series have different patient populations, varying doses of radiation and targets, a variety of assessment methods, and differing follow-up.
• Seventy to 80% of patients are pain free in the short term, although up to 50% may relapse.
• Side effects include facial dysesthesia (up to 12%), corneal irritation, vascular damage, hearing loss, and facial weakness, varying with the dose plan and target area.
• Follow-up is short, and uncertainty persists about possible late complications of radiation therapy.

CONCLUSIONS

MVD is the only surgical option that allows for obtaining long-term pain relief while avoiding any sensory disturbance. In our opinion, it still remains the treatment of choice for all patients with drug-resistant typical TN. Old age and central demyelinations do not constitute absolute contraindications to this kind of surgery. Any age-related statistically significant difference in incidence of surgical complications has been demonstrated. In addition, although the results of MVD in patients affected by MS (as well as the results of percutaneous methods) (11) are less satisfactory, approximately 40% of MS-TN patients was found completely pain free at long-term follow-up. Because sensorial deficits can be far from negligible and well tolerated in some patients treated with lesive procedures, our policy is to delay as much as possible destructive surgery. When these procedures cannot be avoided, PMC should be first proposed because it is easy to perform with infrequent general morbidity, especially on the trigeminal sensitivity. In cases requiring more aggressive treatments because of recurrent pain, TRZ can be used. The use of radiosurgery is still under investigation, and further studies are required to clarify its role in the treatment of TN.

Radiosurgery experts propose their radiation therapy methodology as an etiological strategy because of modifications of axonal conductivity at the root entry zone level. However, radiosurgery should be considered as a lesional method until this hypothesis will be definitively confirmed. In MS patients, unfortunately, both MVD and lesioning procedures cannot prevent pain recurrence because of MS-related evolving demyelination. Thus, new treatments aiming to modulate the activity of central trigeminal pathways should be investigated to improve the quality of life of these patients, refractory to all available surgical and medical therapies. Chronic deep brain stimulation at the thalamic-hypothalamic level might be an interesting option that is at present under evaluation at our institution.