Trigeminal Neuralgia Secondary to Meningiomas and Vestibular Schwannoma Is Improved after Stereotactic Radiosurgery: A Systematic Review and Meta-Analysis Open Access

Idiopathic trigeminal neuralgia (TN) is frequently generated by nerve dysfunction due to the vascular compression of the trigeminal nerve root as it enters the brainstem. Microvascular decompression remains the standard treatment modality, whenever feasible [1, 2]. Other alternatives are radiofrequency lesioning, glycerol rhizotomy, balloon percutaneous compression [3], and radiation techniques. It has been acknowledged that stereotactic radiosurgery (SRS) alleviates idiopathic TN with very few side effects [4, 5].

TN secondary to tumors is encountered in up to 6% of patients with facial pain syndromes and is considered to be associated with tumors affecting the trigeminal nerve [6-9]. These tumors are confined to the TN pathways and might include cavernous sinus, Meckel’s cave, cerebellopontine angle, petrous apex, petroclival, and can be benign or malignant, e.g., meningiomas, trigeminal schwannomas, vestibular schwannomas (VS), epidermoid cysts, metastasis, and so on [6-8, 10]. Meningiomas and schwannomas are the most common tumors to cause secondary TN. Medication alone has a short-term effect, with a high failure rate, ranging from 63 to 100% [8]. Microsurgical resection is the most effective means of relieving pain, but some tumors cannot be completely excised and resection is associated with morbidity and mortality [6-8]. Also, in patients with an excessive surgical risk, including the elderly, or those with significant comorbidities, open microsurgical resection might not be the treatment of choice. Moreover, if the tumors fill in the trigeminal cistern, transovale needle placement is not feasible. In this context, and during the past 2 decades, SRS has been considered a valuable alternative to microsurgical resection.

Here, we focus on the outcomes of SRS for secondary TN generated by benign tumors, particularly the most common ones, meningiomas and VS [6, 7]. Only a few studies have been dedicated to the treatment of tumor-related facial pain with radiosurgery. In the literature, different targeting strategies and a heterogeneity of clinical results are reported. Hence, there is a need to better understand outcomes and further improve the selection of patients for this indication.

A PubMed search was performed for entries between January 1990 and December 2019 using the following query guidelines: ([trigeminal AND (radiosurgery OR Gamma Knife)] AND [vestibular schwannoma]; or combinations with [meningioma], [skull-base], [benign skull-base]). The beginning of the 1990s was chosen as the starting date because prior to this, there were only a few studies published on meningiomas and VS in general that included SRS. Inclusion criteria required that each article be a peer-reviewed clinical study or a case series of meningiomas and VS treated with SRS, independent of the device (Linear Accelerator-Linac; Gamma Knife, GK) used. As such, case reports, non-English studies, conference papers, and abstracts were not included. Some studies advocated targeting the tumor (most [9, 11-20]), while others advocated targeting the tumor and the nerve (the minority [21, 22]). If a study reported a vast majority of meningiomas and VS and only 1–2 cases of other pathologies, it was included in our analysis. Exclusion criteria were: cases treated with fractionated radiotherapy (unless this was the minority in a larger series) [12], series reporting 2 procedures (only reported by Huang et al. [14]), or series focusing on trigeminal schwannomas which are reputed to have different results and only 1 target (the tumor) [23]. Were also excluded series reporting malignant lesions, as these have a different radiobiology and α-to-β ratio [24]. The article selection is illustrated in Figure 1, which includes studies detailed in Tables 1 and 2. Two separate reviewers applied the inclusion criteria to the PubMed search result; there were no disagreements. Moreover, 4 separate reviewers applied the exclusion criteria to the articles that remained, and we finally included 13 series [9, 11-22].

This study was performed in accordance with the published Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [25]. In extracting data from these studies, we paid attention to the classical outcomes usually described after SRS for idiopathic TN as per individual patient (in studies individually reporting the outcomes), i.e., pain relief after SRS, pain recurrence, the maintenance of pain relief, and complications [4, 5]. Freedom from pain was evaluated whenever possible using Barrow Neurological Institute (BNI) pain intensity scores of I = no trigeminal pain and no medication; II = occasional pain, but not requiring medication; III = some pain, but adequately controlled with medication; IV = some pain, but not adequately controlled with medication; and V = continuous severe pain or no pain relief [26].

With regard to the radiology aspect, the probability of an incorrect diagnosis based only upon neuroimaging findings in the absence of histology was previously evaluated by Flickinger et al. [27]as 1.4%. All studies analyzed here stated that all cases had typical imaging features of meningiomas and VS. One series, by Kano et al. [15], applied a unique protocol where all patients received an intravenous dose of 20–40 mg of methylprednisolone after radiosurgery.

Due to the high variation in study characteristics, a statistical analysis using a binary random-effects model (DerSimonian-Laird method) was performed. We used OpenMeta (Analyst) from the Agency for Healthcare Research and Quality.

Weighted summary rates were determined using meta-analytical models. Testing for heterogeneity was performed for each meta-analysis. Pooled estimates using meta-analytical techniques were obtained for all the individual outcomes previously described in the same section.

Pain relief after SRS was associated with a higher maximum dose and single-branch involvement [19].

Pain relief after SRS BNI I was reported in 52/103 cases, i.e., a rate of 50.5% (range 36–65.1; I² = 58.4%, p_{heterogeneity} = 0.035, p < 0.001; Fig. 2a). Pain relief after SRS BNI I–IIIb was encountered in 189/228 cases, i.e., a rate of 83.8% (range 77.8–89.8; I² = 32.69%, p_{heterogeneity} = 0.147, p < 0.001; Fig. 2b).

Pain relief after SRS BNI I–IIIb was encountered in 29/36 cases, i.e., a rate of 83.7% (range 62.3–90.8%; I² = 71.1%, p_{heterogeneity} = 0.063, p < 0.001; Fig. 2c). Time to pain relief is reported in Table 2. Time to decrease in tumor volume was considered longer than time to TN response (18.6 vs. 5.3 months) [11].

Tumor shrinkage after SRS was not necessarily associated with a better clinical outcome, BNI I–IIIb [19].

Maintenance of BNI I was reported in 42/129 cases, for a rate of 36.4% (range 20.1–52.7; ^{^2} = 75.8%, p_{heterogeneity} = 0.002, p < 0.001; Fig. 3a). Moreover, pain relief after SRS BNI I–IIIb was encountered in 29/70 cases, for a rate of 41.2% (range 29.8–52.7; I² = 0%, p_{heterogeneity} = 0.628, p < 0.001; Fig. 3b).

Kim et al. [21] reported 13/15 (86.7%) BNI I–IIIb in their series.

Recurrence is described in 70/205 patients for series focusing on tumor targeting, for a rate of 34.7% (range 21.7–47.6; I² = 76.86%, p_{heterogeneity} < 0.001, p < 0.001; Fig. 4a).

Complications were reported in 24/174 cases in series focusing on tumor targeting, for a rate of 12.6% (range 6.3–18.8; I² = 35.95%, p_{heterogeneity} = 0.154, p < 0.001; Fig 4b). Kim et al. [21] reported 4/15 (26.7%) cases targeting both the tumor and the nerve.

In the current literature, there is a gap of knowledge with regard to potential improvements in secondary TN related to meningiomas or VS. Here, we analyzed series independently of their targeting policy (i.e., the tumor itself vs. the tumor and the nerve vs. the nerve only). Pain relief after SRS was reported as BNI I in 50.5% (range 36–65.1%) and BNI I–IIIb in 83.8% (range 77.8–89.8%). There was no significant difference across series regarding the outcomes with tumor targeting and tumor and nerve targeting. Recurrence was described in 34.7% (range 21.7–47.6; tumor targeting). Maintenance of BNI I was reported in 36.4% (range 20.1–52.7) and maintenance of BNI I–IIIb in 41.2% (range 29.8–52.7%; tumor targeting series). When both the nerve and tumor were targeted, 1 series reported BNI I–IIIb in 86.7% at the last follow-up. Complications were encountered in 12.6% (range 6.3–18.8%; tumor targeting series); however, the rate was much higher, at 26.7%, in the only series that reported them after targeting both the nerve and the tumor. The most common complication was facial numbness.

TN secondary to benign skull-base tumors is a specific case scenario. Of interest are both tumor control and symptomatic improvement. Microsurgical resection has the advantage of a decrease in both the tumor volume and the mass effect on the nerve. In tumor-related cases, Barker et al. [6] reported a series with a mean postoperative follow-up of 9 years, with frequent and long-lasting pain relief, i.e., complete, as high as 81% at 10 years after surgery, and partial, 4%. In the surgical series, new cranial nerve neuropathies were reported in 6.25–19.2% of cases [6, 8, 10]. In selected cases, SRS, and gamma knife surgery in particular, has been shown to be safe and effective in the long term in large cohorts of skull-base meningiomas [28-31] and VS [32-34].

With regard to the clinical aspects, several issues warrant further discussion. Bullitt et al. [7] suggested that the specific trigeminal site of the tumor influences symptoms and signs, i.e., tumors that are peripherally placed cause atypical TN, those in the middle fossa cause severe atypical TN, and those in the posterior fossa cause typical TN. The mechanism that generates pain is considered to be direct trigeminal root compression or vascular compression secondary to a displacement by the tumor [6].

While the tumor is the primary target, the mechanism of action of SRS in secondary TN remains largely undiscovered. However, Régis et al. [35] suggested that there are different mechanisms of action for the relief of secondary tumor-related facial pain by SRS, and they performed experimental studies to confirm this. Chang et al. [11] suggested that reduced abnormal electrical transmission might explain such changes. Huang et al. [14] suggested that the volume of demyelination of the nerves in tumors is probably higher than in idiopathic TN, as is the volume of radiation during treatment. Thus, the dose threshold for pain relief may be lower in the tumor condition. In fact, the maximum dose prescribed for idiopathic TN (70–90 Gy) is much higher [4, 5] than that prescribed for tumors (26–32 Gy) [28, 29]. Some authors include neuropathic pain cases, and report that patients are not responsive to SRS [17].

An open question is whether tumor response, particularly tumor decrease, after SRS is linked with alleviating TN. Chang et al. [11] reported 1 case in which pain disappeared and 3 cases with pain reduction in a series of 27 cases, even though there were no changes in tumor volume during follow-up. Huang et al. [14] suggested that tumor shrinkage may help with the relief of tic douloureux. In sum, tumor shrinkage was associated with pain relief in some of the series [14], while other authors did not find any associations [15]. Some series suggested that pain recurs more often in patients with meningiomas than in those with VS [13].

Overall, complications were rare. The most common complication was facial numbness. Beside numbness, Chang et al. [11]reported transient tinnitus and diminished hearing (in the VS context). Interestingly, there were no cases reported of corneal anesthesia or anesthesia dolorosa. Other complications included balance problems [18], masseter weakness [19], abducens palsy [19], and ipsilateral hemispheric stroke [19]. Kim et al. [21] reported a series of single-session SRS targeting of both nerve and tumor, with no further facial weakness, altered taste function, or hearing and balance difficulties. Of note, hearing loss seems to be encountered more with VS than with meningiomas of the skull-base, as reported in recent studies [36, 37].

In the near future, treatment should be tailored to a patient’s specific needs, as suggested by other groups using SRS (Cyberknife, Accuray) which have reported their algorithm in dealing with tumor-related facial pain, e.g., the Stanford group, using a dose regimen of 18–21 Gy in 3 fractions [38]. Specific issues remain to be developed by further studies: local control might not necessarily improve symptomatology, pain relief might occur after a latency period, and benign tumors respond slow to irradiation.

The most difficult question to answer, from currently available data, is: What is the appropriate target? Moreover, should this be followed by a second targeting, in the same session, or a long time after the first? Would a boost to the nerve be necessary if the tumor is the target?

This analysis had some limitations. The first is the retrospective nature of the included series. The second is related to the fact that some reports included other pathologies besides meningiomas and VS, although in a minor proportion; this might have influenced some of the outcomes. The third is that not all the authors used the BNI or other standard scales for reporting. The fourth is that several studies included malignant or biologically aggressive cases. The fifth, which applies to all TN studies, was the examination of the final outcome, which should be based on clinical examinations rather than telephone interviews. The sixth is that preoperative analysis revealed neuropathic pain in some of the cases. The seventh is related to the imbalance in the number of patients undergoing the different targeting techniques, with a particularly a low number of cases in which both the nerve and tumor were targeted.

SRS for benign tumors, such as meningiomas and VS, is associated with favorable clinical outcomes, although not as much as SRS for idiopathic TN. The current literature illustrates heterogeneity in targeting policy and reporting of the results.

Though targeting both the nerve and the tumor seems to achieve better long-term results, the rate of complications is much higher and the number of patients treated is limited. Future reports should focus on potential randomization between different targeting techniques and uniformity in reporting the results.

The authors thank Lausanne University Hospital and the Faculty of Biology and Medicine, University of Lausanne.

The research presented here was conducted in accordance with the World Medical Association Declaration of Helsinki, the appropriate guidelines for human studies, and animal welfare regulations including the Animal Research: Reporting of in vivo Experiments (ARRIVE) guidelines. It was approved by the appropriate institutional review bodies. For this type of study, formal ethics committee approval is not necessary.

C.T. is a scientific advisor for Elekta Instruments, AB, Sweden, but with no relation to this paper. The other authors report no conflicts of interest.

C.T. gratefully acknowledges receipt of a “Young Researcher in Clinical Research Grant” (“Jeune Chercheur en Recherche Clinique”) from the Faculty of Biology and Medicine, University of Lausanne, and the Lausanne University Hospital (CHUV).

Each author (1) made substantial contributions to the conception or design of the work, or to the acquisition, analysis, or interpretation of data for the work; (2) participated in drafting the work or revising it critically for important intellectual content; (3) approved the final version to be published; and (4) agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

I.P.-F. and J.R. contributed equally as first authors and N.R. and C.T. as senior authors.