An oral nutritional supplement with TRPM8 agonists as a cooling flavor improved swallowing function in post-stroke patients with oropharyngeal dysphagia Authors: Noemí Tomsen, Daniel Españó, Adrian Nuñez-Lara, Pere Clavé Published: 2026 ABSTRACT Post-stroke patients with OD (PSOD) show impaired oropharyngeal sensory function, impaired safety during voluntary swallow (VS) and a critical reduction of spontaneous swallowing frequency (SSF), increasing the risk of aspiration of oropharyngeal secretions. Our aim was to evaluate the effect of oropharyngeal sensory stimulation on the VS and SSF in chronic PSOD patients using a cooling sensation flavoring (TRPM8 agonists) in an oral nutritional supplement (ONS) drink and comparing it with a control ONS. We performed a prospective, crossover, interventional and open-label clinical study on 50 PSOD patients. Patients were randomized into two groups, which differed from each other according to the order in which the ONS were administered: (1) Group A (n = 25): 1st visit, cooling sensation (active) + 2nd visit, control; and (2) Group B (n = 25): 1st visit, control + 2nd visit, active. The swallowing function was evaluated pre and post administration of each product using the volume-viscosity swallowing test (V-VST) for the VS, including the prevalence of impaired safety and efficacy, and SSF (swallows/minute). A saliva sample before and after the administration of control and active ONS’ was collected to quantify substance P (SP) and calcitonin gen-related peptide (CGRP) by using ELISA. We observed that mean basal SSF in PSOD was low (0.29 ± 0.21 swallows/min). PSOD patients showed a significant increase in SSF (51.6%, p < 0.0001), a 22% (p = 0.0449) reduction in the prevalence of impaired safety during VS and a 15% (p = 0.0453) increase in the concentration of salivary SP after the administration of the active ONS. In contrast, no significant changes were observed after the administration of the control product. The order of administration did not affect the results. In conclusion, the TRPM8 compound included in the active ONS facilitates sensory perception, enhancing the peripheral release of SP and improving SSF and safety of swallow during VS in PSOD, which suggests both cortical, and brainstem enhancement of swallow responses.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1038/s41598-026-58808-0. Introduction Oropharyngeal dysphagia (OD) is a frequent complication after stroke. Post-stroke OD (PSOD) is a dynamic condition, some patients may show spontaneous recovery during the acute phase whereas others may maintain signs and symptoms into the chronic phase and even develop new ones. However, it has been described that the prevalence of PSOD remains very high in both the acute and chronic stages, with prevalences of 39.7% (34% with signs of impaired safety) on admission (acute) and 41.7% (19.4% with signs of impaired safety) at three months follow-up1. The pathophysiology of chronic PSOD is characterized by impairment in both the biomechanics and neurophysiology of swallowing leading to alterations in the two main swallow motor patterns: the involuntary and the voluntary swallow response2–4. The main biomechanical impairment in voluntary swallow is characterized by a slow oropharyngeal swallow response (OSR). We found the most significant changes to be the delay in the time to laryngeal vestibule closure (LVC) and to upper esophageal sphincter opening (UESO), increasing the prevalence of penetrations or aspirations of part of the bolus into the respiratory tract during the pharyngeal phase5,6. In addition, we found time to LVC was the critical biomechanical event that protects the respiratory tract during swallowing, as LVC in healthy subjects is 160 ms and a cutoff time > 340 ms in older, post-stroke and dementia patients predicts unsafe swallows1,7,8. Impaired efficacy of swallow is mainly associated with decreased bolus velocity and kinetic energy due to weak tongue propulsion forces5. The neurophysiological swallow response has been much less studied. The aging process increases the pharyngeal sensory threshold to electrical stimulus and impairs latency and amplitude of the pharyngeal sensory evoked potential (pSEP) peaks. Consequently, older patients with OD have decreased pharyngeal sensitivity with a delay in the conduction and integration velocity of pharyngeal sensory stimuli. In addition they present a significant reduction in the concentration of the neuropeptides substance P (SP) and calcitonin gene-related peptide (CGRP) in saliva9,10. We found similar results in patients with chronic post-stroke OD. In this case, patients showed an increased latency and reduced amplitude of pSEPs peaks, but also a loss in the symmetry of the potentials and their cortical representation when comparing brain hemispheres3. In addition, the pharyngeal motor-evoked potentials were symmetric in 73.3% of post-stroke patients with chronic OD without physiologic hemispheric dominance and with a significant reduction in cortical excitability of efferent pathways2. Besides having impaired cortical control of voluntary swallowing (VS), post-stroke patients also show a significant reduction in spontaneous swallowing frequency (SSF)4,11,12 which is a rhythmic motor behavior controlled by the brain stem13. Taken together, all these results support a relevant pathophysiological role of impaired oropharyngeal sensory function in both cortical and brain stem centers controlling both voluntary and spontaneous swallowing and a potential therapeutic target for PSOD patients. Current OD treatment for PSOD is mainly based on compensatory strategies that improve swallow safety14–17. One of these strategies is the use of thickening products, which improve the safety of swallow by increasing the viscosity of fluids. There are several studies that demonstrate the therapeutic effect of thickeners in reducing the prevalence of penetrations and aspirations, and their use has been correlated with fewer cases of respiratory infections, aspiration pneumonia and hospital readmissions6,16,17. In recent years, new neurorehabilitation treatment strategies for OD have been developed such as peripheral (electrical or chemical stimulation) or central stimulation (transcranial direct current stimulation or repetitive transcranial magnetic stimulation). Peripheral stimulation appears to be the most promising strategy, as transcutaneous electrical stimulation (TES) in PSOD patients has been shown to lower the prevalence of unsafe swallows, reduce time to LVC, enhance bolus kinetics and decrease the need for fluid thickening, with therapeutic effects lasting up to one year post-treatment18,19. The human oropharynx and larynx is innervated by afferent fibers of the lingual branch of the cranial nerves V and IX, the chorda tympani of the cranial nerve VII, the pharyngeal branch of the cranial nerves IX and X and the laryngeal branch of the cranial nerve X20. In addition, there are several structures and receptors located in the oropharyngeal mucosa, among which are the transient receptor potential (TRP) channels that have important roles in the integration of sensory stimuli. TRP receptors are widely expressed in several human tissues21, including the human oropharynx, with differential location and expression levels: TRPV1 are found in the mucosa epithelial cells and sensory nerves located below the basal lamina or innervating the mucosa; TRPA1 are found in fibroblast cells and sensory nerves located below the basal lamina22; and TRPM8 are located in sensory nerve fibers below the basal lamina23. Regarding their expression, TRPV1 receptors show a gradient expression, with the highest concentration on the tongue, intermediate in the pharynx and low in the epiglottis. TRPA1 receptor expression, however, is lower than that of TRPV1 in all mentioned locations and with similar expression levels between them22. These receptors are stimulated with various TRP agonists24, and their activation induces the secretion of the neuropeptides SP and CGRP and increases sensory perception and improves swallowing function4,10,25. Ebihara et al. observed a decrease in the swallow reflex latency in stroke and dementia patients in a dose-dependent manner when drops of capsaicin were applied directly into the pharynx26. Some years later, in acute clinical trials, capsaicin was found to reduce prevalence of both safety and efficacy impairments, when administered in a concentration of 150µM to the bolus in patients with OD, by shortening the time to LVC and increasing bolus velocity and cortex excitability19,27–29. The stimulation of TRPV1 with capsaicin has also been found to increase SSF in post-stroke patients by enhancing the transmission of sensory inputs to the brain stem12, improve the efficacy of swallowing by increasing the pharyngeal contractile integral and the time to UESO activation30 and induce the secretion of SP and CGRP, increasing its concentration in patients’ saliva30–33. Regarding the TRPA1 agonist, piperine, the olfactory stimulation with black pepper oil and oral stimulation with piperine within the bolus (10− 4-10− 3 M) also significantly reduced prevalence of impaired safety of swallow in patients with OD and improved the time to LVC34,35. The use of menthol (TRPM8 agonist), either administered into the pharynx (10− 4-10− 2M) or mixed with the bolus (10− 3-10− 2M), reduced the prevalence of safety impairment by improving the swallow response24,36. Finally, other TRP agonists whose therapeutic effect has been evaluated are the TRPA1 agonists, cinnamaldehyde, citral and the combined TRPA1-TRPM8 agonist citral-isopulegol. We observed that the stimulation of TRPA1 receptors, specially with cinnamaldehyde, significantly improved the prevalence of safe swallows, the time to LVC and to UESO, and brain activation37. The aim of this study was to evaluate the therapeutic effect of the pre-thickened oral nutritional supplement (ONS) drink, Nutilis Complete® Cooling Sensation (Danone Nutricia, Zoetermeer, The Netherlands) which contains a mixture of TRPM8 agonists, on signs of impaired safety and efficacy during voluntary swallow, spontaneous swallowing frequency and the concentration of salivary neuropeptides in chronic post-stroke patients with OD. Methods Study design and patients This was a randomized, crossover, interventional, and open-label clinical study conducted on patients with chronic post-stroke oropharyngeal dysphagia (OD) referred to the Dysphagia Unit at Mataró University Hospital. Each patient acted as his own control and attended two visits. At each visit, the swallowing function was assessed, using the volume-viscosity swallow test (V-VST), and spontaneous swallowing frequency (SSF) before and after the administration of a pre-thickened ONS drink with Cooling Sensation (CS) or control. In addition, a sample of saliva was collected to quantify the salivary neuropeptides. The study was conducted in accordance with the consensus ethical principles outlined in international guidelines, including the Declaration of Helsinki and Council for International Organizations of Medical Sciences International Ethical Guidelines, Good Clinical Practice guidelines and all applicable laws and regulations. This included the relevant local legislation of the country where the research was carried out, with priority given to the laws that offer the highest level of protection to the participants. The study protocol was reviewed and approved by the Ethics Committee of the Consorci Sanitari del Maresme (EC code: 45/23) and registered in the ClinicalTrials.gov (NCT06250426) in January 2024. The patients were randomized into 2 groups: Group 1: 1st visit, stimulation with CS, 2nd visit, stimulation with control; and Group 2: 1st visit, stimulation with control, 2nd visit, stimulation with CS. The washing out period between the two visits was at least 7 days. Patients were instructed to avoid menthol-like compounds during this period. The main inclusion criteria were to be more than 18 years old, diagnosed with OD secondary to stroke more than 3 months from the onset, with impaired safety and/or efficacy of swallow during the V-VST and able to follow the protocol. Exclusion criteria included being pregnant, a life expectancy of less than 3 months, a diagnosis of OD prior to stroke, dementia with a Global Deterioration Scale higher than 3, allergy to the investigation product, participation in another interventional study in the month prior to the inclusion or being a current user of pre-thickened ONS with CS. All patients signed the informed consent before their inclusion in the study. The recruitment phase and completion of study visits were conducted between October 2023 and January 2025. Accepting an alpha risk of 0.05 and a beta risk of 0.1 in a two-sided test, a minimum of 22 participants per group are necessary to recognize a difference greater than or equal to 0.5 swallows/min as statistically significant with a standard deviation of 0.6. Anticipating a drop-out rate of 30%, a final sample of 50 subjects (25 per group) were required. Randomization Patients who met all the inclusion criteria and none of the exclusion criteria were randomized in one of the two groups. The randomization was performed using GraphPad QuickCalcs® software (https://www.graphpad.com/quickcalcs/) through a list of random numbers. Study products Both products (control and active product) were Nutilis Complete® (Danone Nutricia, Utrecht, The Netherlands), the control being Nutilis Complete strawberry flavor, while the active product was Nutilis Complete Cooling Sensation lemon mint flavor. The composition and rheological characteristics between both study products are similar (viscosity range 385–705 mPa.s at 50s− 1) except for the inclusion of a combination of TRPM8 agonists in the active product. Experimental design A total of 50 ml of either active or control, depending on randomization and the visit, was administered orally. The stimulation period consisted of 4 boluses of 10 ml administered every 5 min. At each of the 5 min intervals, patients were asked to perform one dry swallow every minute. Finally, just prior to performing the second V-VST, 10 ml of the product was administered (Fig. 1). Fig. 1Study design. V-VST: volume-viscosity swallowing test; SSF: spontaneous swallowing frequency. Study design. V-VST: volume-viscosity swallowing test; SSF: spontaneous swallowing frequency. Swallowing assessment Volume-viscosity swallowing test The Volume-Viscosity Swallowing Test (V-VST) is a validated effort test in which boluses of increasing volume and difficulty are administered to check for clinical signs of impaired efficacy (impaired labial seal, oral residue, pharyngeal residue and piecemeal deglutition) and safety (changes in voice quality, cough or a decrease in oxygen saturation ≥ 3%) in each swallow38. It starts with 250 mPa·s viscosity and increasing volumes from 5 ml, to 10 ml and 20 ml boluses in a progression of increasing difficulty. When patients have completed the 250 mPa·s series without major symptoms of aspiration, a less safe liquid (< 50 mPa·s) viscosity series is assessed also with boluses of increasing difficulty (5, 10 and 20 ml). Finally, a safer 800 mPa·s viscosity series (5, 10 and 20 ml) is assessed in the same way. If the patient presents signs of impaired safety at 250 mPa·s viscosity, the series is interrupted, the < 50 mPa·s viscosity series omitted, and the safer 800 mPa·s viscosity series is assessed. If the patient presents signs of impaired safety at < 50 mPa·s viscosity, the liquid series is interrupted, and the 800 mPa·s series is assessed. The V-VST was performed twice per visit, one at the start of the visit and the other at the end. The thickened fluids used were obtained by mixing 2.4 g (250 mPa·s) or 4.2 g (800 mPa·s) of Nutilis Clear (Danone Nutricia, Zoetermeer, The Netherlands) in 100 ml of water in both the control and active group. Spontaneous swallowing frequency Spontaneous swallowing frequency (SSF) was measured with surface neck electromyography (sEMG) and accelerometry for 10 min, in which the patient was asked to breathe normally and to remain calm and quiet. sEMG electrodes were placed over the digastric-mylohyoid complex and an omnidirectional accelerometer was positioned over the cricothyroid cartilage to assess the number of swallows per minute of the recordings using the AcqKnowledge 4.4 software (BIOPAC Systems, Inc., Goleta, CA, USA)4,12. Saliva collection and quantification of salivary neuropeptides Saliva samples were collected, using the Salivette® technique-putting a swab under the tongue for 5 min-and stored at -80 °C until their analysis. The concentration of salivary neuropeptides substance P (SP) and calcitonin gene-related peptide (CGRP) were determined by using 2 specific commercial enzyme-linked immunosorbent assay (ELISA) kits: substance P parameter assay kit (Ref. KGE007; R&D systems, USA) and CGRP (Human) ELISA kit (Abnova, Taiwan). A total of two samples were collected at each visit (one before and one after the stimulation). Statistical analysis Although the study was open-label, the researcher who performed the data analysis was blinded in order to avoid bias. Given that a one-week washing period was introduced between visits, and after verifying that there were no significant differences between the pre intervention values of each of the visits after performing an unpaired t-test for continuous data and a chi-square test for categorical data, the data were analyzed using two different methods: (1) within groups: pre vs. post comparison per visit within the same group using a paired t-test (continuous data) and a chi-square test (categorical data); and (2) between groups: pre vs. post comparison independently of the administration order of control or active product by using an unpaired t-test (continuous data) and a chi-square test (categorical data). Finally, a difference-in-difference (DiD) analysis was performed for those variables that showed significant post-intervention differences to estimate the causal effect of the intervention. DiD was calculated as (post active-pre active) - (post control-pre control). A DiD > 0 o < 0 was interpreted as a positive or negative effect of the active product, respectively. Statistical significance was assessed using the interaction term between intervention (active vs. control product) and time (pre vs. post measurement) in a two-way ANOVA. Continuous variables were expressed as mean ± standard deviation while categorical data was expressed as relative or absolute frequencies. Non-parametric tests were used on non-Gaussian variables. All statistical tests were performed with GraphPad Prism 9.2.0 (GraphPad Software, San Diego, CA, USA). Significant differences were considered when p-value < 0.05. Results Study population A total of 50 chronic PSOD patients were finally included in the study (67.86 ± 11.81 years, 50% women). They presented mild-to-moderate ischemic stroke associated with mild impairment of functional status and with 66% of patients showing clinical signs of unsafe swallow. No significant differences in socio-demographic and clinical basal data were found between study groups, as shown in Table 1. All patients completed the two arms of the study protocol. Table 1Basal clinical and socio-demographic data. Values are expressed as mean ± SD, except for those specifically stated.TotalGroup AGroup Bp-valuen502525Age (years)67.86 ± 11.8166.84 ± 11.9468.88 ± 11.830.5468Gender (% female)505248> 0.9999Stroke type (%)Ischemic75.67080.950.3219Hemorrhagic12.21014.29Undetermined12.2204.76mRS2.31 ± 1.602.04 ± 1.702.58 ± 1.470.2364NIHSS3.33 ± 3.183.08 ± 3.333.58 ± 3.060.3691Swallow efficacy impairment (%)100100100> 0.9999Swallow safety impairment (%)666864> 0.9999SSF (swallows/min)0.29 ± 0.220.36 ± 0.240.23 ± 0.190.0592n: sample size; mRS: modified Rankin Scale; NIHSS: National Institute of Health Stroke Scale; SSF: spontaneous swallowing frequency. Table 1Basal clinical and socio-demographic data. Values are expressed as mean ± SD, except for those specifically stated.TotalGroup AGroup Bp-valuen502525Age (years)67.86 ± 11.8166.84 ± 11.9468.88 ± 11.830.5468Gender (% female)505248> 0.9999Stroke type (%)Ischemic75.67080.950.3219Hemorrhagic12.21014.29Undetermined12.2204.76mRS2.31 ± 1.602.04 ± 1.702.58 ± 1.470.2364NIHSS3.33 ± 3.183.08 ± 3.333.58 ± 3.060.3691Swallow efficacy impairment (%)100100100> 0.9999Swallow safety impairment (%)666864> 0.9999SSF (swallows/min)0.29 ± 0.220.36 ± 0.240.23 ± 0.190.0592n: sample size; mRS: modified Rankin Scale; NIHSS: National Institute of Health Stroke Scale; SSF: spontaneous swallowing frequency. Basal clinical and socio-demographic data. Values are expressed as mean ± SD, except for those specifically stated. n: sample size; mRS: modified Rankin Scale; NIHSS: National Institute of Health Stroke Scale; SSF: spontaneous swallowing frequency. Effect on spontaneous swallowing frequency In both groups, a significant increase in SSF was observed only when patients received the active product: from 0.36 ± 0.24 swallows/min to 0.54 ± 0.27 swallows/min (p = 0.0030) in Visit 1 for Group A (50.56% increase), and from 0.30 ± 0.26 swallows/min to 0.46 ± 0.26 swallows/min (p = 0.0008) in Visit 2 for Group B (52.78% increase) (Supplementary Fig. 1). When analysing the results of all patients together, significant differences were still only observed when patients took the active product, going from an SSF of 0.33 ± 0.25 swallows/min to 0.50 ± 0.27 swallows/min (p < 0.0001) (51.55% increase). No significant differences in SSF were observed for the control product (Fig. 2). Fig. 2Spontaneous swallowing frequency (SSF) pre and post the control or active intervention ****: p < 0.0001. Spontaneous swallowing frequency (SSF) pre and post the control or active intervention ****: p < 0.0001. Effect on voluntary swallow during the volume-viscosity swallowing test Only those patients that received the active product in Visit 1 (Group A) showed a significant reduction in the prevalence of pharyngeal residue (from 92% to 64%, p = 0.0374) (Supplementary Table 1), but no differences were observed at Visit 2 (Supplementary Table 2) or when patients received the control product at either visit (Table 2). Likewise, when the data from all 50 patients were analyzed together according to whether they received the control or the active product, a significant reduction in the prevalence of safety impairment signs (from 60% to 38%, p = 0.0449) (Fig. 3), specifically in the prevalence of voice changes due to contact of the fluid with the vocal folds (from 48% to 24%, p = 0.0213) and a tendency in the reduction of the prevalence of pharyngeal residue (from 80% to 62%, p = 0.0769) were observed only when patients received the product with the cooling sensation compounds (Table 2). Table 2Prevalence (%) of clinical signs of safety and efficacy of voluntary swallowing pre and post control and active intervention.Control productActive productPrePostp-valuePrePostp-value% Impaired efficacy9898> 0.99999898> 0.9999% Oral residue68600.532374580.1389% Pharyngeal residue70600.401980620.0769% Piecemeal deglutition9088> 0.999982780.8031% Impaired safety66600.679060380.0449% Cough2224> 0.999918220.8031% Voice change5250> 0.999948240.0213% Oxygen desaturation64> 0.999924> 0.9999 Table 2Prevalence (%) of clinical signs of safety and efficacy of voluntary swallowing pre and post control and active intervention.Control productActive productPrePostp-valuePrePostp-value% Impaired efficacy9898> 0.99999898> 0.9999% Oral residue68600.532374580.1389% Pharyngeal residue70600.401980620.0769% Piecemeal deglutition9088> 0.999982780.8031% Impaired safety66600.679060380.0449% Cough2224> 0.999918220.8031% Voice change5250> 0.999948240.0213% Oxygen desaturation64> 0.999924> 0.9999 Prevalence (%) of clinical signs of safety and efficacy of voluntary swallowing pre and post control and active intervention. Fig. 3Effect of control and active products on the prevalence (%) of patients with signs of l impaired safety of swallow. *: p < 0.05. Effect of control and active products on the prevalence (%) of patients with signs of l impaired safety of swallow. *: p < 0.05. Effect on concentration of salivary neuropeptides No significant differences were observed in the secretion of the salivary neuropeptide SP in either group when Visit 1 (active: from 111.98 ± 60.69 pg/ml to 122.21 ± 74.99 pg/ml, p = 0.2705; control: from 83.68 ± 37.99 pg/ml to 139.30 ± 103.71 pg/ml, p = 0.1605) and Visit 2 (control: from 130.33 ± 69.59 pg/ml to 129.03 ± 75.64 pg/ml, p = 0.6194; active: from 94.22 ± 53.87 pg/ml to 117.22 ± 68.31 pg/ml, p = 0.0937) were analyzed separately (Supplementary Fig. 2). In contrast, when analysing the results of all 50 patients together, a significant 15% increase in the concentration of this neuropeptide was observed when patients received the active product (from 119.30 ± 87.84 pg/ml to 131.12 ± 97.45 pg/ml, p = 0.0453) but not when received the control (from 108.30 ± 60.90 pg/ml to 116.92 ± 74.92 pg/ml, p = 0.6922) (Fig. 4a). Fig. 4Salivary (A) substance P (SP) and (B) calcitonin gene-related peptide (CGRP) concentration pre and post control or active intervention. *: p < 0.05. Salivary (A) substance P (SP) and (B) calcitonin gene-related peptide (CGRP) concentration pre and post control or active intervention. *: p < 0.05. Regarding the neuropeptide CGRP, no significant differences were observed in the secretion of this neuropeptide after administration of the control or the active product, either when analyzed by Visit 1 (active: from 51.54 ± 37.37 pg/ml to 38.90 ± 29.99 pg/ml, p = 0.1440; control: from 75.18 ± 82.03 pg/ml to 105.01 ± 107.08 pg/ml, p = 0.1250) and Visit 2 (control: from 32.04 ± 24.07 pg/ml to 76.40 ± 80.49 pg/ml, p = 0.0714; active: from 44.43 ± 46.04 pg/ml to 54.40 ± 70.92 pg/ml, p = 0.4609) (Supplementary Fig. 3) or by total effect (control: from 36.07 ± 34.41 pg/ml to 58.93 ± 61.31 pg/ml, p = 0.0695; active: from 46.76 ± 40.94 pg/ml to 36.47 ± 34.32 pg/ml, p = 0.0624) (Fig. 4b). Causal effect of the intervention According to DiD analysis, there was a significant positive causal effect of the active product on the SSF (DiD = 0.098, p = 0.032) (Fig. 5a) and a significant negative causal effect on prevalence of safety impairment signs (DiD=-16, p = 0.044) (Fig. 5b). Regarding salivary SP, a non-significant positive causal effect was observed (DiD = 4.67, p = 0.747) (Fig. 5c). Fig. 5Causal effect representation of (A) spontaneous swallowing frequency (SSF), (B) prevalence of impaired safety and (C) salivary substance P (SP) concentration. Causal effect representation of (A) spontaneous swallowing frequency (SSF), (B) prevalence of impaired safety and (C) salivary substance P (SP) concentration. Discussion The results of the present study suggests that oropharyngeal sensory stimulation by using TRPM8 agonists could improve both the voluntary and the involuntary/spontaneous swallow response in post-stroke patients with OD, as it has been observed a significant reduction in the prevalence of safety impairment clinical signs at V-VST and increased SSF and salivary SP following an acute stimulation with a cooling sensation flavoring. The natural history and potential complications of a patient with OD as a consequence of a cerebrovascular accident is well described. The prevalence of OD in post stroke patients is very high, whether they are in the acute (45%) or chronic phase (41%). Regarding the complications, OD is an independent risk factor for prolonged hospital stay, respiratory infection and in-hospital mortality during the admission, and almost 40% of patients present nutritional risk39. The prevalence of these complications remains high at three months and one year after the stroke, as 11% of patients experience respiratory infections, impaired functional capacity (25%), severe nutritional complications and mortality higher in patients with OD compared to those without39,40. These data demonstrate that post-stroke OD is strongly associated with severe clinical outcomes, reinforcing the need for new therapeutic strategies that can induce the improvement of swallowing response. To date, there is still no active treatment for post-stroke OD. The physiopathology of OD in chronic PSOD patients is characterized by an impairment in both the biomechanical and the neurophysiological swallowing response. At the neurophysiological side patients with post-stroke OD not only presented an impairment in the afferent or sensory pathway, by showing a loss in the symmetry of the pSEPs and a severe delay of the conduction and integration velocity of the sensory stimuli3, but also an impairment in the efferent or motor pathway, by presenting a loss of the dominance of the pharyngeal motor cortex and a reduction in cortical excitability3,41. At the biomechanical side of the voluntary swallowing, PSOD patients presented prolonged total duration of swallowing, a weak propulsion (0.14 J, 10 cm/s) and a slow reconfiguration, especially of the time to LVC which a delayed time higher than 340 ms indicate a very high risk of unsafe swallows and aspirations1. Regarding the involuntary swallowing, those post-stroke patients with OD had a very reduced SSF (0.25 swallows/min)4. Taken altogether, there is a huge need to develop new specific therapeutic strategies that can stimulate the neurophysiology of swallowing to improve both motor swallowing outputs as a consequence. To date, different central and peripheral neurostimulation strategies have been investigated as potential treatments for OD in post-stroke patients. All these treatments have shown high potential for improving swallowing response in this phenotype of patients with OD, with TRP agonists showing the greatest potential, although further studies are still needed42–44. Nowadays, the most used treatment for this phenotype of patients with OD are the thickening products. The therapeutic effect of thick fluids on voluntary swallowing has been widely described. On the one hand, it is known that the effect on the prevention of unsafe swallowing is shear viscosity-dependent, determining that the optimal viscosities to protect the highest percentage of patients were 250 and 800 mPa·s. On the other hand, higher levels of viscosity can significantly impair the efficacy of swallowing, especially by increasing the prevalence of oral and pharyngeal residue. Few studies show that increased fluid viscosity decreases the time to LVC in patients with OD16,17,45. This effect may be secondary to the activation of the multiple mechanoreceptors found in the mucosa and the somatosensory neurons that innervate it, facilitating the oropharyngeal swallowing response as a consequence46–48. However, other studies evaluating the effect of thickening products on neurophysiological response have not observed that increased viscosity had a significant effect on its rehabilitation37,49. We did not observe any significant effect when patients received the control product, either on voluntary or involuntary swallowing. This result is also consistent with the description of thickened fluids as a mostly compensatory treatment. In contrast, the acute stimulation with natural TRP agonists showed significant therapeutic effects on biomechanical and neurophysiological swallowing response in older and post-stroke patients with OD. From a biomechanical point of view, capsaicin (TRPV1), piperine (TRPA1/V1), menthol (TRPM8), cinnemaldehyde-zinc (TRPA1) and citral (TRPA1) significantly reduced the time to LVC and to UESO, reducing the prevalence of unsafe swallows as a consequence24,28,29,33,34,50. In addition, capsaicin also increased the SSF in PSOD patients12 and the contractile integral the UES activation and relaxation times30. Furthermore, when comparing their therapeutic potential with compensatory strategies, we observed that some agonists (such as piperine or citral) had similar therapeutic potential to that of thickening agents but only TRP agonists (whether TRPV1, TRPA1 or TRPM8) also improved the swallowing response24. The results of the present study showed that the acute administration with a product including TRPM8 agonists significantly reduced by up to 22% the prevalence of clinical signs of unsafe swallows with a significant negative causal effect, specifically the voice change sign, suggesting the reduction in the prevalence of penetrations/aspirations, in post-stroke patients with OD. Although only the clinical signs of voluntary swallowing were assessed, this result agrees with previous studies using videofluoroscopy, where the acute effect of oral stimulation with menthol significantly reduced the prevalence of patients with safety impairment signs between 4 and 37%, depending on the concentration (1 or 10 mM, respectively). This effect is probably secondary to the shortening of the timing of the oropharyngeal swallowing response observed after the stimulation, especially of the time to LVC24,50, but further studies using an instrumental assessment are needed to describe the effect on swallowing physiology. Regarding the effect on involuntary swallowing, we observed a significant increase of 50% in SSF only when patients received the active compound containing the TRPM8 agonists with a significant positive causal effect. By using a rat model, it is known that the activation of TRPM8 and TRPV1 channels in laryngopharyngeal regions innervated by the superior laryngeal nerve facilitates the involuntary swallowing response51. In addition, in a previous study performed with older patients it has been described that cold-sensitive stimulation using menthol shortened the latent time of swallowing reflex52. A similar effect in post-stroke patients with OD had already been described for capsaicin (TRPV1 agonist), which increased the SSF in this phenotype of patients after an acute stimulation12. Together with our findings, we can suggest that TRP agonists are promising sensory stimulants to improve the swallowing response in PSOD. Finally, our results showed that the TRPM8 agonist combination included in the active product induced a significant increase in the secretion of the salivary neuropeptide SP, but with a non-significant causal effect and without inducing any effect on the secretion of CGRP. The results obtained for CGRP are not inconsistent with previous studies suggesting that TRPM8 channel activation may exert anti-inflammatory effects through TRPV1 inhibition and the subsequent reduction in the secretion of proinflammatory neuropeptides (including CGRP)53,54. However, our findings do not directly confirm TRPM8-mediated modulation of CGRP release. In contrast, the increased secretion of SP goes in consonance with previous studies describing the use of TRP agonists as swallowing stimulants. It has been observed that the use of TRPV1 agonists (mainly capsaicin) significantly increased the secretion of SP and facilitates, as a consequence, the swallowing response30,33,55. Although the role of TRPM8 in swallowing function has not been specifically investigated, existing evidence indicates that activation of TRPM8 channel by cold temperatures or cooling compounds induces Ca+ 2 influx and increases sensory afferent excitability through both direct neuronal activation and indirect epithelial mechanisms56–58. TRPM8-mediated stimulation can promote the realease of adenosine triphosphate (ATP) from epithelial cells, whichs acts as paracrine neurotransmitter on purinergic P2X/P2Y receptors expressed on adjacent afferent nerve terminals59,60. This ATP-dependent purinergic signalling pathway enhances the excitability of sensory fibers and facilitates the release of SP59,61, amplifying the sensory input transmission within brain cortex and brainstem62,63. Since it has been described that a low concentration of salivary SP is directly associated with reduced pharyngeal sensitivity10 and decreased SSF64, this increase in SP would be beneficial in reducing the prevalence of penetrations/aspirations in PSOD patients by enhancing sensory perception. In addition, recent evidence from in vitro and animal model studies suggests that ATP may also directly contribute to involuntary swallowing initiation through purinergic signalling pathways65–68, raising the possibility that TRPM8 activation could influence swallowing through both ATP-dependent and SP-mediated mechanisms. Further studies in humans are needed to confirm tha mechanism by which TRPM8 activation modulated involuntary swallowing control. In summary, these findings suggest that administration of an oral nutritional supplement containing TRPM8 agonists may activate afferent sensory nerve endings, leading to the secondary release of the neuropeptide SP. This mechanism could facilitate the transmission of sensory inputs to the brainstem and sensory cortex, enhancing their neuroreactivity and increasing SSF and safety of swallowing as a consequence. Although the results are encouraging, the study has several limitations. First, we have only evaluated the therapeutic effect after acute administration. Because the product is an oral nutritional supplement, it is designed for frequent and prolonged use over time. Future studies in which consumption of Nutilis Complete Cooling Sensation is maintained would be necessary to determine its long-term effect. It would also be interesting to study whether the addition of a TRPM8 agonist to the supplement’s composition has any additional effect on the nutritional status of patients. Finally, in the present study we focused on chronic PSOD. It would be interesting to study how other patient phenotypes (specially older and dementia patients) in different stages of evolution (acute, subacute, or chronic) respond to this nutritional supplement. Finally, despite the good psychometric properties of the V-VST, confirmation of the effect on voluntary swallowing using instrumental assessment techniques will be required in future studies. It would also be of interest to further investigate the neurophysiological effects of this product through comprehensive evaluation of both the afferent and efferent pathways involved in the neurophysiological swallowing response. In conclusion, Nutilis Complete Cooling Sensation significantly increased the SSF by 50%, reduced the prevalence of clinical safety impairment signs by 22% and increased the secretion of salivary SP by 15% in chronic post-stroke with OD. The TRPM8 compound included in Nutilis Complete Cooling Sensation could enhance the neurophysiological swallowing response controlled by the brain cortex and brainstem through the facilitation of oropharyngeal sensory perception. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1 Supplementary Material 2 Supplementary Material 2 Supplementary Material 3 Supplementary Material 3 Supplementary Material 4 Supplementary Material 4