Chapter 99 Stereotactic Radiosurgery for Trigeminal Neuralgia
Trigeminal neuralgia (TN) has been defined as sudden, paroxysmal, recurring stabbing pains of one or more branches of the trigeminal nerve that are generally unilateral.1 Classic TN is typically secondary to vascular compression or is idiopathic. Patients may recognize different pain triggers, such as chewing, talking, washing, shaving, smoking, or tooth brushing. Between pain episodes, patients tend to be asymptomatic, although a minority of patients continue to experience background pain.2,3 Many interventions exist for the treatment of TN. Application of stereotactically delivered radiation to treat TN has become a useful, minimally to noninvasive treatment option.
Historical Background
Ionizing radiation for the treatment of TN dates to within a few years of when x-rays were discovered, but it eventually grew out of favor.4 When stereotactic radiosurgery (SRS) was invented, the use of radiation for TN was revisited. In 1971, Leksell reported on two patients with classic TN whom he treated in 1953 with a single dose of ionizing radiation targeted on the gasserian ganglion and root entry zone.5 Results were unsatisfactory until the target was moved from the gasserian ganglion to the proximal trigeminal root near the pons.6 The effects of stereotactically delivered radiation at the root entry zone produces focal axonal degeneration and necrosis of the trigeminal nerve without affecting the gasserian ganglion, which is thought to be the mechanism resulting in pain relief in TN patients.7 Use of SRS, usually with the gamma knife machine, became widely acceptable in 1996 after a multi-institutional study demonstrated favorable results.8 Other technologies delivering stereotactic radiation to treat TN have also been reported. In 2003, Romanelli et al. reported the first use of the CyberKnife to treat TN.9 In the same year, Smith et al. reported the first series of patients with TN treated with linear accelerator (LINAC) radiosurgery.10
Indications
The initial treatment for TN is medical therapy, with carbamazepine being the gold standard and oxcarbazepine, a keto derivative of carbamazepine, a suitable alternative with more tolerable side effects. SRS is recommended for patients with medically refractory TN who are not surgical candidates, since microvascular decompression (MVD) has the highest rates of pain control in patients who are surgical candidates.11 SRS is regarded as the safest minimally invasive therapy for TN and is particularly useful in patients who are older than 70 years, have significant medical comorbidities, or decide not to undergo MVD.4,12 However, pain relief is not immediate and may take months.13 Ideally, SRS should be performed in patients who can undergo magnetic resonance imaging (MRI) for treatment planning. For patients with MRI-incompatible pacemakers or defibrillators, computed tomography (CT) with contrast cisternography can be used.14
As with other surgical treatments for TN, patients with typical TN symptomatology fare the best after SRS. Therefore, an accurate diagnosis is critical, because SRS may not be the best treatment option for patients with atypical features. For example, TN pain related to multiple sclerosis can be satisfactorily treated with SRS but not with the same success as classic TN.15 SRS can be used as a first-time interventional treatment and as repeat or salvage treatment for those who continue to have pain following SRS or other procedures.
Gamma Knife Technique
Gamma knife radiosurgery is a frame-based technique that delivers radiation from a fixed cobalt source. Planning and treatment occur the same day in a single session. On the day of treatment, the first step involves the placement of the stereotactic Leksell frame under local anesthesia. Typically, four points (two anterolateral and two occipital) are chosen to ensure appropriate fixation of the frame (Fig. 99-1). The insertion points are injected with a mixture of 2% lidocaine with sodium bicarbonate for local anesthesia. Bupivacaine can be injected concurrently for long-term analgesia. The screws are then inserted into the frame and skull. The appropriate screw should be selected so that it sits flush with the frame. If a screw is too long, it will lie outside of the frame, create additional artifact during the MRI, and possibly interfere with the treatment beams. It is also critical that the anterior screws be applied while the eyes of the patient are closed. If the screws are applied while the patient’s eyes are open, complete eye closure may not be possible, making it uncomfortable for the patient and risking corneal ulceration.
After application of the Leksell frame is completed, a 1-mm-slice MRI with gadolinium is obtained. T1, T2, and fast imaging employing steady-state acquisition (FIESTA) sequences allow trigeminal nerve identification and its relationship to adjacent blood vessels to be examined.16 The target is a single, 4-mm isocenter positioned at the trigeminal nerve root entry zone, 2 to 4 mm from the junction of the nerve and the pons8,13,17 (Fig. 99-2). Maintenance of this buffer zone between the nerve and the pons keeps the radiation delivered to the brain stem between 20% and 30% of the isodose.8 Final dose selection and calculation of the dosage delivery plans are done in conjunction with a radiation oncologist and a physicist.
Once the planning is complete, treatment can proceed. The patient is placed on the treatment bed, and the collimator helmet is affixed (Fig. 99-3). Once the target coordinates for the beam have been confirmed, the treatment commences. After the treatment is complete, the Leksell frame is taken off by sequentially removing the screws opposite each other. The pin site defects are covered with gauze and antibiotic ointment.
Linear Accelerator
Other frame-based technologies have emerged, such as LINAC radiosurgery. Similar to gamma knife radiosurgery, frame placement on the patient, treatment planning, and the radiosurgical treatment session take place in the same day. The dose fallout areas have been shown to be comparable to those of gamma knife.18
Cyberknife Technique
An MRI FIESTA sequence with 1.25-mm thin cuts and a CT with 1.25-mm thin cuts should be obtained. The MRI is used to localize and target the trigeminal nerve root entry zone.19–21 The CT images are used as a template for localization during the treatment. The images from CT and MRI are fused together to combine the two data sets for treatment. The CT/MRI fusion technique has become standard practice at centers utilizing CyberKnife for the treatment of TN and has obviated the need for obtaining a CT-cisternogram to localize and target the trigeminal nerve root entry zone.20 As a result, treatment of TN with CyberKnife has become a noninvasive procedure.
Once the trigeminal nerve and its root entry zone have been identified, a segment of the nerve is targeted for treatment. The segment should be 2 to 3 mm from the dorsal root entry zone to minimize the dose delivered to the brain stem. At this distance, the brain stem receives between 30% and 50% of the maximum treatment dose20 (Fig. 99-4). Likewise, extension into the gasserian ganglion in Meckel’s cave should be avoided. To minimize hot spots and heterogenous radiation dosages delivered to the nerve, a nonisocentric plan is used.
Once the planning is complete, the patient is placed supine on the treatment bed and kept in place using a custom-made thermoplastic mask. During the treatment session, orthogonal cranial x-rays are taken between every three to five treatment shots. The x-ray films are analyzed with reference to the previously loaded CT data and compared during the session to detect patient movement. When patient movement has occurred, the target coordinates are recalculated and the movements of the robotic arm movements are adjusted accordingly.22,23 Approximately 30 x-rays are taken during the treatment session, which is roughly equivalent to the radiation dose delivered during a routine CT scan.20 Using a 1.25-mm-slice CT scan in the data set, the accuracy of the target is roughly within 1.1 mm.24
Ideal Dosage and Targeting
Several factors affect the outcomes and side effects from treatment of TN with radiosurgery. These include the specific portion of the nerve treated, the dose delivered, and the length of nerve treated. Targeting of the gasserian ganglion was shown by multiple groups to reduce pain in a select number of patients but had poor long-term results.25,26 Improved short- and long-term results have been seen when the proximal nerve is targeted shortly after its takeoff from the pons. In this segment, the source of myelin of the nerve changes from oligodendrocytes (central) to Schwann cells (peripheral). It has been hypothesized that this zone is more sensitive to radiation.17 At the same time, minimizing the radiation delivered to the brain stem is of paramount importance.
In the first multi-institutional study conducted by Kondziolka et al., the effect of different dosages delivered to the trigeminal nerve using the gamma knife was examined. A significantly higher number of patients (72% vs. 9%) had complete pain relief with treatment doses greater than 70 Gy compared to those treated with doses less than 70 Gy.4,8