Artificial Intelligence to Predict Surgical Outcomes and Assess Pain Neuromodulation in Trigeminal Neuralgia Subjects

Artificial Intelligence to Predict Surgical Outcomes and Assess Pain Neuromodulation in Trigeminal Neuralgia Subjects

Artificial Intelligence to Predict Surgical Outcomes and Assess Pain Neuromodulation in Trigeminal Neuralgia Subjects

Trigeminal neuralgia (TN) is the most common cause of facial pain. Medical treatment is the first therapeutic choice whereas surgery, including Gamma Knife radiosurgery (GKRS), is indicated in case of pharmacological therapy failure. However, about 20% of subjects lack adequate pain relief after surgery. Virtual reality (VR) technology has been explored as a novel tool for reducing pain perception and might be the breakthrough in treatment-resistant cases. The investigators will conduct a prospective randomized comparative study to detect the effectiveness of GKRS aided by VR-training vs GKRS alone in TN patients. In addition, using MRI and artificial intelligence (AI), the investigators will identify pre-treatment abnormalities of central nervous system circuits associated with pain to predict response to treatment. The investigators expect that brain-based biomarkers, with clinical features, will provide key information in the personalization of treatment options and bring a huge impact in the management and understanding of pain in TN.

Trigeminal Neuralgia, Trigeminal Nerve Diseases, Virtual Reality, Artificial Intelligence, Radiosurgery, Magnetic Resonance Imaging

Single centre, randomized controlled trial, assessor blinded. The investigators plan to enroll 50 TN patients with indications to undergo surgical treatment. Fifty TN participants will be randomly allocated into two groups: individuals in the experimental group (GKRS-VR) will undergo GKRS followed by immersive VR-based sensorimotor neuromodulation training, whilst control group (CT) will undergo only GKRS. A group of 50 age- and sex-matched controls will also be recruited.

Physiotherapist, neurologists, neuropsychologists, radiologist and biomedical engineering assessing the patients are blinded to group allocation.

GKRS-VR group will undergo GKRS and neuromodulation based on Virtual Reality sensorimotor rehabilitation using an immersive system

Age- and sex-matched healthy subjects recruited to compare neuropsychological,clinical and structural/functional magnetic resonance imaging characteristics at baseline.

One month after radiosurgery, GKRS-VR subjects will receive neuromodulation treatment twice a week (30' per time), for 8 consecutive weeks followed by 9 months of remote rehabilitation. GKRS-VR group will perform neuromodulation based on VR sensorimotor rehabilitation using an immersive system developed ad-hoc. Treatments will include voluntary mimic movements reinforced by VR augmented feedback, consisting in the reflection of a real-time avatar animation, multiplied when needed, associated with different targets to reach based on specific facial gestures. The execution of motion multiplier on avatar will be allowed in order to personalize treatments, emphasizing a graded exposure to pain perception. In addition, the investigators will develop a dedicated mobile-application for remote rehabilitation with the aim of prescribing VR augmented feedback home exercises and immersive VR pain neuromodulation.

Control Group will undergo only radiosurgery with no rehabilitation and they will perform only clinical evaluation at study entry, after three months and one year.

Changes in time in the Numeric rating scale (NRS): a pain screening tool, commonly used to assess pain severity at that moment in time using a 0-10 scale, with zero meaning "no pain" and 10 meaning "the worst pain imaginable"

Changes in time in the Barrow Neurological Institute (BNI) pain intensity scale to assess the level of pain in patients with trigeminal neuralgia. BNI values I-III were considered to indicate good outcomes whereas BNI values IV and V indicate poor response to treatment.

The McGill Pain Questionnaire (MPQ) can be used to evaluate a person experiencing significant pain. It can be used to monitor the pain over time and to determine the effectiveness of any intervention. The users are presented with a list of 78 words in 20 sections that are related to pain.The users mark the words that best describe their pain (multiple markings are allowed). Among the words, sections of these words signify different components of pain, namely, Sensory (sections 1-10), Affective (sections 11-15), Evaluative (section 16), and Miscellaneous (sections 17-20).

The Pain Catastrophizing Scale (PCS) assesses the extent of catastrophic thinking. It is a 13-item scale, with a total range of 0 to 52. Higher scores are associated with higher amounts of pain catastrophizing.

The Central Sensitisation Inventory (CSI) is a self-report outcome measure designed to identify patients who have symptoms that may be related to central sensitisation.The CSI consists of two parts: Part A includes 25 questions related to common central sensitivity syndromes symptoms.Part B determines if the patient has been diagnosed with certain central sensitivity syndromes disorders or related disorders, such as anxiety and depression.

The Tampa Scale for Kinesiophobia (TSK) is a 17-item questionnaire that quantifies fear of movement. Individual item scores range from 1-4, with the negatively worded items (4,8,12,16) having a reverse scoring (4-1). The 17 item TSK total scores range from 17 to 68 where the lowest 17 means no or negligible kinesiophobia, and the higher scores indicate an increasing degree of kinesiophobia.

Craniofacial Pain and Disability Inventory (CF-PDI) can be used in research and clinical practice for the assessment of patients with craniofacial pain. It is a 21-item measure and patients give their answers using a 4-point Likert scale scoring from 1 (no problem) to 4 (maximum problem). The scale presents two factors, named Pain and disability (14 items) and Jaw functional status (7 items). The scores of the responses are added with higher scores indicating greater problems

The Penn Facial Pain Scale (Penn-FPS) was originally developed in order to fully assess the impact of trigeminal neuralgia pain on patients' health-related quality of life. It is a 7-item scale, with a total range of 0 to 10.

A series of reflex responses after electrical stimulation of the ophthalmic division (R1 and R2) and maxillary/mandibular division (SP1 and SP2) of the trigeminal nerve

Functional MRI will be acquired on a 3T scanner during the performance of a task, in which participants will be asked to observe facial gestures during activities of daily living or specific movements different for each subject depending on which action triggers the facial pain.

Inclusion Criteria: diagnosis of TN according to the International Classification of Headache Disorders, third edition (ICHD-3) criteria; age >18 years; indication to undergo radiosurgical treatment; willingness and ability to comply with scheduled visits and other trial procedures. Exclusion Criteria: Any person unable to lie still within the environment of the MRI scanner for the required period to perform the study and those where MRI scanning is contraindicated; metal implants, pacemaker, etc.; Pregnancy or breastfeeding; Any significant psychiatric disease; Use of illicit drugs; Brain pathology shown by brain MRI and/or neurophysiological examination; Any person unable to understand and follow the instructions of the investigators; Any other condition according to the Investigator would make the subject unsuitable for the study.

Barzaghi LR, Pompeo E, Albano L, Del Vecchio A, Mortini P. Gamma Knife radiosurgery for cluster-tic syndrome unresponsive to medical treatment: illustrative case. J Neurosurg Case Lessons. 2021 Jul 26;2(4):CASE2191. doi: 10.3171/CASE2191. eCollection 2021 Jul 26.

Barzaghi LR, Albano L, Scudieri C, Gigliotti CR, Nadin F, Del Vecchio A, Mortini P. Gamma Knife Radiosurgery for Trigeminal Neuralgia: Role of Trigeminal Length and Pontotrigeminal Angle on Target Definition and on Clinical Effects. World Neurosurg. 2020 Oct;142:e140-e150. doi: 10.1016/j.wneu.2020.06.147. Epub 2020 Jun 26. Erratum In: World Neurosurg. 2021 Mar 4;:

Albano L, Agosta F, Basaia S, Castellano A, Messina R, Parisi V, Barzaghi LR, Falini A, Mortini P, Filippi M. Alterations of brain structural MRI are associated with outcome of surgical treatment in trigeminal neuralgia. Eur J Neurol. 2022 Jan;29(1):305-317. doi: 10.1111/ene.15105. Epub 2021 Sep 24.

Filippi M, Agosta F, Scola E, Canu E, Magnani G, Marcone A, Valsasina P, Caso F, Copetti M, Comi G, Cappa SF, Falini A. Functional network connectivity in the behavioral variant of frontotemporal dementia. Cortex. 2013 Oct;49(9):2389-401. doi: 10.1016/j.cortex.2012.09.017. Epub 2012 Oct 24.

Filippi M, Basaia S, Canu E, Imperiale F, Meani A, Caso F, Magnani G, Falautano M, Comi G, Falini A, Agosta F. Brain network connectivity differs in early-onset neurodegenerative dementia. Neurology. 2017 Oct 24;89(17):1764-1772. doi: 10.1212/WNL.0000000000004577. Epub 2017 Sep 27.

Agosta F, Spinelli EG, Marjanovic IV, Stevic Z, Pagani E, Valsasina P, Salak-Djokic B, Jankovic M, Lavrnic D, Kostic VS, Filippi M. Unraveling ALS due to SOD1 mutation through the combination of brain and cervical cord MRI. Neurology. 2018 Feb 20;90(8):e707-e716. doi: 10.1212/WNL.0000000000005002. Epub 2018 Jan 24.

Basaia S, Agosta F, Cividini C, Trojsi F, Riva N, Spinelli EG, Moglia C, Femiano C, Castelnovo V, Canu E, Falzone Y, Monsurro MR, Falini A, Chio A, Tedeschi G, Filippi M. Structural and functional brain connectome in motor neuron diseases: A multicenter MRI study. Neurology. 2020 Nov 3;95(18):e2552-e2564. doi: 10.1212/WNL.0000000000010731. Epub 2020 Sep 10.

Basaia S, Agosta F, Diez I, Bueicheku E, d'Oleire Uquillas F, Delgado-Alvarado M, Caballero-Gaudes C, Rodriguez-Oroz M, Stojkovic T, Kostic VS, Filippi M, Sepulcre J. Neurogenetic traits outline vulnerability to cortical disruption in Parkinson's disease. Neuroimage Clin. 2022;33:102941. doi: 10.1016/j.nicl.2022.102941. Epub 2022 Jan 19.

Basaia S, Agosta F, Wagner L, Canu E, Magnani G, Santangelo R, Filippi M; Alzheimer's Disease Neuroimaging Initiative. Automated classification of Alzheimer's disease and mild cognitive impairment using a single MRI and deep neural networks. Neuroimage Clin. 2019;21:101645. doi: 10.1016/j.nicl.2018.101645. Epub 2018 Dec 18.

Alemanno F, Houdayer E, Emedoli D, Locatelli M, Mortini P, Mandelli C, Raggi A, Iannaccone S. Efficacy of virtual reality to reduce chronic low back pain: Proof-of-concept of a non-pharmacological approach on pain, quality of life, neuropsychological and functional outcome. PLoS One. 2019 May 23;14(5):e0216858. doi: 10.1371/journal.pone.0216858. eCollection 2019.

Emedoli D, Arosio M, Tettamanti A, Iannaccone S. Virtual Reality Augmented Feedback Rehabilitation Associated to Action Observation Therapy in Buccofacial Apraxia: Case Report. Clin Med Insights Case Rep. 2021 Feb 15;14:1179547621994579. doi: 10.1177/1179547621994579. eCollection 2021.