# Evaluating the diagnostic role of magnetic resonance imaging in trigeminal neuralgia, hemifacial spasm, and glossopharyngeal neuralgia: A controlled blinded study

**Authors:** Roberta Bonomo, Giulio Bonomo, Guglielmo Iess, Marco Paolo Schiariti, Jacopo Falco, Mario Stanziano, Francesco Restelli, Elio Mazzapicchi, Morgan Broggi, Ikrame Labiad, Francesco Acerbi, Giuseppe M.V. Barbagallo, Paolo Ferroli

PMC · DOI: 10.1016/j.bas.2025.105607 · Brain & Spine · 2025-09-25

## TL;DR

MRI alone is not reliable for diagnosing nerve compression in facial pain conditions without clinical context.

## Contribution

First study to evaluate MRI in all three HDS using blinded raters and healthy controls.

## Key findings

- MRI accuracy for detecting NVC improves with more clinical information provided to raters.
- Blinded MRI assessments showed low interrater agreement and poor diagnostic reliability.
- NVC can be present in asymptomatic individuals and absent in symptomatic patients.

## Abstract

Hyperactive dysfunction syndromes (HDS)—including trigeminal neuralgia, hemifacial spasm, and glossopharyngeal neuralgia—are typically associated with neurovascular compression (NVC) at the cranial nerve root entry/exit zone. MRI is widely employed to detect NVC, although its diagnostic reliability remains controversial, as NVC may also be observed in asymptomatic individuals, while in some symptomatic patients, radiological evidence can be absent.

Can preoperative MRI reliably identify NVC in HDS patients undergoing microvascular decompression (MVD), and how does clinical information affect diagnostic accuracy?

We retrospectively analyzed preoperative 3 T MRIs from 42 HDS patients treated with MVD (2013–2023) and compared them with scans from age-matched healthy controls. Three experienced raters evaluated the images under five conditions with varying levels of clinical information, to assess how context influenced the detection of NVC.

Diagnostic accuracy increased in parallel with the amount of clinical information available. Under fully blinded conditions, correct identification of NVC remained below 50 %. Sensitivity improved when the affected nerve was disclosed (77.7 %) and further increased with full clinical context (83.3 %). Sensitivity decreased to 23.8 % when control scans were introduced without the raters’ awareness but rose to 88.9 % when controls were acknowledged. Interrater agreement remained low (k = 0.2–0.3) across all settings.

MRI findings alone may be insufficient to confirm or exclude NVC in HDS. Interpretation should always be integrated with clinical data, and surgical decisions should not be based solely on imaging, particularly in symptomatic patients with negative scans.

•First study assessing MRI in all HDS using blinded raters and healthy controls•MRI diagnostic accuracy improves when clinical information is available•NVC may be absent on MRI in symptoms or present in asymptomatic individuals•Blinded MRI assessment shows low interrater agreement in detecting NVC•MRI alone is not sufficient to guide surgical indication in HDS patients

First study assessing MRI in all HDS using blinded raters and healthy controls

MRI diagnostic accuracy improves when clinical information is available

NVC may be absent on MRI in symptoms or present in asymptomatic individuals

Blinded MRI assessment shows low interrater agreement in detecting NVC

MRI alone is not sufficient to guide surgical indication in HDS patients

## Linked entities

- **Diseases:** trigeminal neuralgia (MONDO:0008599), glossopharyngeal neuralgia (MONDO:0016372)

## Full-text entities

- **Diseases:** HDS (MESH:D001289), trigeminal neuralgia (MESH:D014277), hemifacial spasm (MESH:D019569), glossopharyngeal neuralgia (MESH:D020435), NVC (MESH:D013901)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC12549520/full.md

## References

20 references — full list in the complete paper: https://tomesphere.com/paper/PMC12549520/full.md

---
Source: https://tomesphere.com/paper/PMC12549520