Posterior fossa morphology and cranial nerve configuration in trigeminal neuralgia: A morphometric magnetic resonance imaging study Authors: Tomoyuki Naito, Keisuke Onoda, Akira Matsuno, Takayuki Shiomi Published: 2026 ABSTRACT Background: Neurovascular compression is considered the principal cause of trigeminal neuralgia (TN), but vascular contact alone does not fully explain disease occurrence. We hypothesized that posterior fossa morphology and cranial nerve spacing contribute to TN susceptibility.Methods: Fifty-six patients were analyzed: 20 with TN, 20 with hemifacial spasm (HFS), and 16 controls. Using three-dimensional magnetic resonance cisternogram/angiogram fusion imaging, five morphometric parameters were defined: (A) frontal-occipital cranial length, (B) posterior fossa depth, (C) anteroposterior diameter of the brainstem between the roof of the fourth ventricle and the ventral pons measured on midsagittal T1-weighted images, (D) distance between the trigeminal and seventh to eighth cranial nerves, and (E) distance between bilateral trigeminal roots. Ratios B/A, D/C, and E/C were calculated.Results: The TN and HFS groups had significantly smaller posterior fossa depth indices than controls (TN vs. control, p < 0.01; HFS vs. control, p = 0.007). The interneural spacing index was significantly smaller in TN and HFS (p < 0.001) and further decreased on the affected side in TN (p = 0.041). No significant difference was observed in bilateral trigeminal spacing.Conclusion: A compact posterior fossa and reduced cranial nerve spacing may predispose individuals to TN. INTRODUCTION Trigeminal neuralgia (TN) is a well-known paroxysmal pain disorder characterized by brief lancinating facial pain.[3] The classical pathophysiological concept attributes TN to neurovascular compression at the root entry zone of the trigeminal nerve.[3,6] However, vascular contact with the trigeminal nerve is frequently observed in asymptomatic individuals,[5] and many patients with TN exhibit only minor or no visible compression on magnetic resonance imaging (MRI).[4,11] This discrepancy suggests that vascular conflict alone may be insufficient to explain disease occurrence and that additional anatomical or structural factors may influence susceptibility.[2,4,5,10,11] Previous morphometric studies have suggested that posterior fossa size and configuration may influence the likelihood of cranial nerve crowding and subsequent neurovascular conflict.[2,4,5,10,11] Among these studies, Zhong et al. performed a detailed morphometric MRI analysis and demonstrated that patients with TN exhibited characteristic posterior fossa anatomical features compared with controls.[11] Their work provided an important methodological framework for the quantitative assessment of posterior fossa morphology in TN. Building upon this approach, the present study additionally incorporated a hemifacial spasm (HFS) cohort, introduced a novel interneural spacing index, and evaluated side-to-side asymmetry within individual patients. A relatively small posterior fossa could result in reduced cerebrospinal fluid space and closer proximity of cranial nerves and vessels. Such anatomical crowding may increase the probability of symptomatic neurovascular compression. In addition to posterior fossa size, the spatial relationship between cranial nerves within the cerebellopontine cistern may also be an important factor. However, few studies have quantitatively evaluated interneural distances in patients with TN. Furthermore, normalization of morphometric parameters to account for individual skull size differences has not been consistently performed. We therefore hypothesized that both posterior fossa compactness and reduced cranial nerve spacing contribute to the pathophysiology of TN. To test this hypothesis, we conducted a morphometric MRI analysis measuring posterior fossa dimensions and cranial nerve configuration and compared findings among patients with TN, HFS, and control subjects. The present study extends the methodology of Zhong et al. by incorporating an HFS comparison cohort, introducing a novel interneural spacing index (D/C ratio), and evaluating side-to-side asymmetry within individual TN patients.[11] MATERIALS AND METHODS Study design and participants This retrospective morphometric study was designed to clarify whether specific anatomical features of the posterior fossa and cranial nerve configuration predispose individuals to TN. The 56 participants were categorized into three groups: 20 patients with TN, 20 with HFS, and 16 control subjects who underwent MRI for hearing-related symptoms (including hearing disturbance, vertigo, or tinnitus) and were confirmed to have no structural abnormalities on neuroimaging. None of the control subjects had TN, HFS, vestibular schwannoma, or other posterior fossa lesions. Their mean ages were 68.7 ± 13.7, 60.5 ± 11.5, and 59.9 ± 14.6 years in the TN, HFS, and control groups, respectively [Table 1]. There were no significant differences in age or sex distribution between the groups (one-way analysis of variance [ANOVA], p = 0.082), nor in sex distribution. Table 1:Patient demographics. Patient demographics. This was a retrospective exploratory study; therefore, no a priori sample size calculation was performed. The sample size was determined by the number of eligible patients who met the inclusion criteria during the study period. Imaging technique A morphometric analysis was performed using the parameters illustrated in Figure 1, which combines sagittal T1-weighted, axial constructive interference in steady state, and 3D cisternogram/angiogram fusion images. Measurement parameters for morphometric analysis. Three-dimensional magnetic resonance cisternogram/angiogram fusion, sagittal T1-weighted, and axial constructive interference in steady state images were used to define five morphometric parameters: (A) Distance between frontal and occipital poles; (B) distance between the posterior margin of the foramen magnum and sinus confluence; (C) anteroposterior diameter of the brainstem between the roof of the fourth ventricle and the ventral pons; (D) shortest distance between the trigeminal nerve root and the seventh to eighth cranial nerves (arrow: Trigeminal nerve; double arrows: Facial-vestibulocochlear nerve complex); (E) shortest distance between the bilateral trigeminal roots. Ratios (B/A, D/C, and E/C) were used for normalization to eliminate variations in skull size and overall cranial volume. Morphometric parameters Five morphometric parameters were defined: A: Fronto-occipital cranial length B: Posterior fossa depth C: Anteroposterior diameter of the brainstem between the roof of the fourth ventricle and the ventral pons measured on midsagittal T1-weighted images D: Distance between the trigeminal nerve and VII–VIII complex E: Distance between bilateral trigeminal roots To reduce the influence of individual skull size variation, ratios were calculated. B/A = Posterior fossa depth index D/C = Interneural spacing index E/C = Bilateral trigeminal spacing index Normalization using the pons diameter was selected as a stable internal reference. Statistical analyses Comparisons among groups were performed using one-way ANOVA followed by Tukey’s post hoc testing. Continuous variables were expressed as mean ± standard deviation. p < 0.05 was considered statistically significant. Interobserver reproducibility was assessed by repeated measurements, and variability was <5%. For comparisons among the TN, HFS, and control groups, one-way ANOVA followed by Tukey’s post hoc test was used to account for multiple pairwise comparisons. Comparisons between affected and unaffected sides within the TN group and between right and left sides within the control group were prespecified paired analyses and were evaluated separately. Measurements were performed by investigators who were aware of the clinical diagnosis and were therefore not blinded to group assignment. Ethics statement The study was approved by the Institutional Review Board of International University of Health and Welfare Narita Hospital (Approval No. 25-CC-041). Because of the retrospective design, informed consent was waived, and an opt-out procedure was implemented. RESULTS Posterior fossa morphology The posterior fossa depth index (B/A) was significantly smaller in the TN and HFS groups than that in the control group, indicating a more compact posterior fossa configuration [Figure 2]. The mean B/A ratios were 0.194 ± 0.031, 0.209 ± 0.027, and 0.238 ± 0.033 in the TN, HFS, and control groups, respectively. A highly significant difference was observed between the TN and control groups (p < 0.001), along with a significant difference between the HFS and control groups (p = 0.007). The difference between the TN and HFS groups was not significant (p = 0.11), although the TN group tended to have lower ratios. These results suggest that both TN and HFS occur in individuals with a relatively tight posterior cranial fossa compared with the general population. Comparison of the posterior fossa depth index (B/A) among groups. The posterior fossa depth index (B/A), defined as the ratio between the distance from the foramen magnum to the sinus confluence (B) and the total cranial length (A), was significantly smaller in both the trigeminal neuralgia (TN) and hemifacial spasm (HFS) groups compared with that in the control group. This indicates that the posterior fossa was more compact in patients with neurovascular compression syndromes. Error bars represent standard deviation. Significance: TN versus control (p<0.01); HFS versus control (p=0.007); TN versus HFS (not significant). Bars represent group mean values, and error bars indicate standard deviations. Sample sizes were TN (n=20), HFS (n=20), and control (n=16). “**” indicates p < 0.01. Interneural distance between the trigeminal and seventh to eighth cranial nerves The normalized interneural distance (D/C) was markedly shorter in the TN and HFS groups than that in the control group [Figure 3a]. The mean D/C ratios were 0.268 ± 0.048, 0.283 ± 0.051, and 0.348 ± 0.052 for the TN, HFS, and control groups, respectively. The difference was highly significant for the TN group versus the control group (p < 0.001) and the HFS group versus the control group (p < 0.001). Although not significant (p = 0.36), patients with TN tended to have lower ratios than did those with HFS, suggesting a closer spatial approximation of the trigeminal and seventh to eighth cranial nerves. This finding implies that both disorders share a background of cranial nerve crowding; however, the degree of proximity was most pronounced in the patients with TN. Interneural distance index (D/C) among groups. (a) Comparison of interneural distance index (D/C) among trigeminal neuralgia (TN), hemifacial spasm (HFS), and control. The interneural distance index (D/C) represents the ratio between the distance from the trigeminal nerve root to the seventh to eighth cranial nerves (D) and the brainstem anteroposterior diameter (C). Both the TN and HFS groups demonstrated significantly shorter D/C ratios than did controls, suggesting reduced spacing between cranial nerves and greater neural crowding at the level of the pons. Error bars represent standard deviation. Significance: TN versus control (p<0.01); HFS versus control (p<0.01); TN versus HFS (not significant). (b) Comparison of interneural distance index (D/C) between the affected and unaffected sides in TN. In patients with TN, the affected side had a significantly lower D/C ratio compared with that on the contralateral (unaffected) side, indicating that the symptomatic side had greater local structural crowding around the trigeminal root entry zone. Error bars represent standard deviation. Significance: p=0.041. (c) Comparison of interneural distance index (D/C) between sides in the control group. In the control group, there was no significant difference in the D/C ratio between the right and left sides, confirming that the asymmetry observed in patients with TN was disease-specific rather than a normal anatomical variation. Error bars represent standard deviation. Significance: ns: Not significant. Bars represent group mean values, and error bars indicate standard deviations (SD). In panels B and C, ipsilateral/affected and contralateral/unaffected measurements were averaged across all subjects within each group. Bars represent group mean values, and error bars indicate standard deviations (SD). Sample sizes were TN (n=20), HFS (n=20), and Control (n=16). “**” indicates p < 0.01. Side-to-side asymmetry in patients with TN In the TN group [Figure 3b], the affected side exhibited a significantly lower D/C ratio than did the contralateral side (0.257 ± 0.049 vs. 0.289 ± 0.051; p = 0.041). This asymmetry indicates that the symptomatic side tended to have greater local structural crowding within the posterior fossa, possibly predisposing it to neurovascular compression and symptom onset. In contrast, the control group [Figure 3c] showed no significant side-to-side differences in the D/C ratio (p = 0.619), suggesting that this asymmetry is a characteristic feature specific to patients with TN. These findings support the hypothesis that TN preferentially develops in anatomically constrained regions, rather than because of random vascular contact. Bilateral trigeminal root spacing The bilateral trigeminal spacing index (E/C) did not differ significantly among the three groups (TN: 1.458 ± 0.12, HFS: 1.528 ± 0.13, control: 1.502 ± 0.11). Pairwise comparisons revealed no significant differences between TN and HFS (p = 0.083), TN and control (p = 0.263), or HFS and control (p = 0.538) [Figure 4]. Although patients with TN showed a tendency toward lower E/C ratios, this trend did not reach statistical significance. Bilateral trigeminal root spacing index (E/C) among groups. The bilateral trigeminal root spacing index (E/C), defined as the ratio between the distance of the bilateral trigeminal nerve roots (E) and the brainstem diameter (C), did not differ significantly among groups. A slight trend toward smaller values was observed in patients with trigeminal neuralgia (TN), suggesting possible mesial displacement of the trigeminal roots in a compact posterior fossa. Error bars represent standard deviation. TN versus hemifacial spasm (HFS) (p=0.083), TN versus Control (p=0.263), HFS versus Control (p=0.538). Bars represent group mean values, and error bars indicate standard deviations. Sample sizes were TN (n=20), HFS (n=20), and control (n=16). DISCUSSION Our morphometric and comparative analyses suggest that TN is not solely a consequence of an isolated vascular offender but frequently occurs in the context of permissive cranial morphology.[1,2,4,5,7,9,10] Specifically, a reduced posterior fossa sagittal dimension (lower B/A ratio) and decreased interneural distances, most notably between the trigeminal root entry zone and the seventh to eighth cranial nerve complex, were characteristic of TN and HFS compared with controls.[2,4,5,10,11] Moreover, in patients with TN, the affected side exhibited a further reduction in the distance between the trigeminal and seventh to eighth cranial nerves, supporting a focal rather than purely bilateral or systemic morphological predisposition in symptomatic disease.[2,4,5,10,11] Normalization and methodological considerations Previous morphometric studies of TN have primarily focused on the size or volume of the posterior fossa.[4,11] However, relatively few studies have examined the spatial configuration among the cranial nerves themselves. The present study expands this perspective by demonstrating that the distance between the trigeminal nerve and the facial-vestibulocochlear nerve complex is significantly reduced in patients with TN. These findings suggest that cranial nerve crowding, rather than posterior fossa size alone, may represent an important anatomical factor contributing to neurovascular compression syndromes. To minimize the influence of individual differences in overall cranial size, all distance measurements were normalized using a stable internal reference, specifically the anteroposterior diameter of the brainstem along the line connecting the roof of the fourth ventricle to the ventral surface (parameter C).[11] This approach allowed for the comparison of relative proportions rather than absolute dimensions and enabled the identification of structural crowding independent of head size or age-related variations. Unlike previous volumetric studies by Hardaway, Cheng, and Wang, the present method relies on simple normalized linear indices that can be readily applied to routine MRI examinations without dedicated volumetric software. Normalization using brainstem length is particularly appropriate because the brainstem is a rigid, centrally located structure whose size remains relatively constant across individuals and serves as an internal control for intracranial morphometry.[11] In addition, the ratio between the distance from the foramen magnum to the sinus confluence (parameter B) and the fronto-occipital cranial length (A) provides a reproducible index of the posterior fossa depth relative to the total cranial length.[2,4,5,10,11] This normalization strategy enhances the sensitivity of detecting subtle differences in the posterior fossa configuration among groups. Mechanistically, a compact posterior fossa may produce subtle downward and anterior displacement of the upper brainstem and cerebellar structures, reducing the free space that normally separates the cranial nerve rootlets and adjacent vasculature.[2,4,5,10,11] In this crowded environment, even modest asymptomatic arterial or venous loops in the larger posterior fossa may impinge on the trigeminal root entry zone with sufficient chronic pulsatile microtrauma or focal demyelination to generate electric paroxysmal pain. This model reconciles two otherwise discordant observations: (i) the relatively high prevalence of radiologic neurovascular contact in asymptomatic individuals[5] and (ii) that many surgical TN specimens show focal nerve pathology, even when a clear compressive vessel is not obvious.[1,7,8] Our findings have several clinical implications. Preoperative morphometric assessments (nerve-to-nerve distance measurements and volumetric posterior fossa analysis) could refine patient selection for microvascular decompression and aid intraoperative planning by identifying patients with a high likelihood of multiple or positional conflicts.[2,4,5,10,11] In patients without a single dominant arterial offender, recognition of an anatomically tight posterior fossa might shift the surgical strategy toward careful exploration of subtle offending vessels or consideration of additional techniques (thorough arachnoid dissection, transpositional padding) that address global tightness rather than a single contact point.[2,4,5,10,11] In addition, morphological metrics might inform prognostication and counseling; patients with marked posterior fossa crowding could theoretically have different recurrence risks or response profiles compared to those whose imaging shows isolated vascular contact in a spacious posterior fossa.[1,8] From a surgical perspective, recognition of posterior fossa crowding and cranial nerve configuration may also help surgeons anticipate complex neurovascular relationships during microvascular decompression. The present study has several limitations. First, the sample size was modest and drawn from a surgical cohort, which could have introduced selection bias toward more severe or refractory cases.[1,8] Second, although our measurements are reproducible within the dataset, they rely on linear indices and two-dimensional proxies for a 3D space; more refined volumetric or surface-based morphometry would better characterize the posterior fossa geometry.[2,4,5,10,11] Finally, causal inference remains limited in cross-sectional imaging studies. The association between tight posterior fossa morphology and TN is compelling; however, longitudinal population studies are required to determine whether these anatomical features precede symptom onset or represent developmental or age-related correlates.[5,9] Although no statistically significant age difference was observed among the groups, age-related anatomical changes in posterior fossa structures may have influenced the morphometric measurements and should be considered when interpreting the results. Third, the control group was not composed of completely healthy volunteers but of individuals who underwent an MRI for hearing-related symptoms and were subsequently confirmed to have no structural abnormalities. Therefore, the B/A and D/C values observed in the control group may not necessarily represent population-normative values. Although no posterior fossa lesions were identified in these subjects, a degree of selection bias cannot be completely excluded. Measurements were not performed in a blinded manner. Therefore, observer bias cannot be completely excluded and should be considered when interpreting the findings. Future work should pursue larger imaging cohorts with longitudinal follow-up; incorporate high-resolution volumetric MRI and diffusion imaging to assess microstructural nerve changes; and explore whether quantified posterior fossa morphology can improve predictive models for TN onset, surgical outcomes, and recurrence.[2,4,5,10,11] Correlative histopathological analyses of resected or biopsied nerve tissues from patients with different morphologies may clarify the mechanistic links between mechanical stress and focal demyelination. Taken together, these findings suggest that TN may be better understood as a disorder arising from the interaction between vascular contact and individual cranial anatomical configuration rather than from vascular compression alone. CONCLUSION Our results indicated that a structurally tight posterior fossa and reduced distances between cranial nerve rootlets were common in patients with TN and may constitute an anatomical predisposition, in conjunction with vascular contact, which facilitates symptomatic nerve dysfunction. The use of normalized morphometric indices, particularly those referenced to brainstem dimensions, provided a reliable means of quantifying posterior fossa crowding independent of head size and enhanced the reproducibility of such anatomical comparisons. Recognizing posterior fossa morphology as a contributing factor shifts the pathophysiological perspective from a binary “vascular contact causes TN” model toward a multifactorial framework in which individual cranial anatomy modulates vulnerability to neurovascular conflict. The incorporation of a targeted morphometric assessment into preoperative evaluations may enhance diagnostic precision, guide surgical strategies, and ultimately improve individualized patient care.