Dysphagia https://doi.org/10.1007/s00455-022-10451-3 ORIGINAL ARTICLE Spontaneous Swallowing Frequency in Post‑Stroke Patients with and Without Oropharyngeal Dysphagia: An Observational Study Marta Alvarez‑Larruy1,2 · Noemí Tomsen1,3 · Nicolau Guanyabens1,2 · Ernest Palomeras2 · Pere Clavé1,3,4 · Weslania Nascimento1 Received: 10 November 2021 / Accepted: 8 April 2022 © The Author(s) 2022 Abstract Oropharyngeal dysphagia (OD) is a frequent complication after stroke (PSOD) that increases morbidity and mortality. Early detection of PSOD is essential to reduce morbidity and mortality in patients with acute stroke. In recent years, an association between reduced spontaneous swallowing frequency (SSF) and OD has been described. Likewise, the reduction of saliva substance P (SP) concentration has been associated with an increased risk of aspiration and a decrease in SSF. In this study we aimed to compare SSF, salivary SP concentration, hydration and nutritional status in post-stroke (PS) patients with and without OD. We included 45 acute PS patients (4.98 ± 2.80 days from stroke onset, 62.22% men, 71.78 ± 13.46 year). The Volume-Viscosity Swallowing Test (V-VST) was performed for clinical diagnosis of OD. SSF/minute was assessed through 10-min neurophysiological surface recordings including suprahyoid-electromyography and cricothyroid-accelerometry. Saliva samples were collected with a Salivette® to determine SP by ELISA. Hydration status was assessed by bioimpedance. Nutritional status was evaluated by Mini Nutritional Assessment Short Form (MNA-sf) and blood analysis. Twenty-seven PS patients (60%) had OD; 19 (40%), impaired safety of swallow. SSF was significantly reduced in PSOD, 0.23 ± 0.18 and PSOD with impaired safety, 0.22 ± 0.18 vs 0.48 ± 0.29 swallows/minute in PS without OD (PSnOD); (both p < 0.005). Nutri- tional risk was observed in 62.92% PSOD vs 11.11% PSnOD (p = 0.007) and visceral protein markers were also significantly reduced in PSOD (p < 0.05). Bioimpedance showed intracellular dehydration in 37.50% PSOD vs none in PSnOD. There were no differences for saliva SP concentrations. SSF is significantly reduced in PSOD in comparison with PSnOD. Acute PSOD patients present poor nutritional status, hydropenia, and high risk for respiratory complications. Keywords Spontaneous swallowing frequency · Stroke · Swallow · Deglutition · Oropharyngeal Dysphagia Introduction Stroke is the second cause of death in the world [1] and Marta Alvarez-Larruy and Noemí Tomsen both contributed is also a leading cause of disability, which is expected to equally to the study. increase in the coming decades [2] with high socioeconomic costs [3, 4]. Oropharyngeal dysphagia (OD) is one of the * Pere Clavé most important causes of disability and mortality in patients pere.clave@ciberehd.org who have suffered a stroke [1, 2, 5, 6]. According to a study 1 Gastrointestinal Physiology Laboratory, Department carried out by our group [7], the prevalence of post-stroke of Surgery, Hospital de Mataró, Universitat Autónoma de (PS) OD on admission was 45%, with up to 38% patients Barcelona, Mataró, Spain showing clinical signs of impaired safety of swallow. The 2 Department of Neurology, Hospital de Mataró, Mataró, Spain most prevalent sign of impaired safety of swallow was 3 Centro de Investigación Biomédica en Red de Enfermedades voice change observed in up to 43.65% [7]. Undetected and Hepáticas Y Digestivas (CIBERehd), Instituto de Salud untreated patients with OD are at elevated risk of prolonged Carlos III, Barcelona, Spain hospitalization, respiratory infection, pneumonia, malnutri- 4 Department of Surgery, Hospital de Mataró, Universitat tion, dehydration, and death [5, 7–12]. Early identification of Autònoma de Barcelona, Carretera de Cirera s/n 08304, post-stroke oropharyngeal dysphagia (PSOD) is fundamental Mataró, Spain 13 Vol.:(0123456789) M. Alvarez-Larruy et al.: Post-stroke spontaneous swallowing frequency and dysphagia to reduce morbidity and mortality during hospital admission identify clinically significant OD with a sensitivity of 96% and after hospital discharge [13, 14]. and a specificity of 68%, and that a 5–10 min sampling win- The volume-viscosity swallow test (V-VST), an effort dow was sufficient to calculate SSF. The authors concluded test that uses boluses of different volumes and viscosities that SSF has high potential to identify PSOD with psycho- to identify clinical signs of impaired efficacy and/or safety metric properties equal or superior to clinical screening pro- of swallow [15], is a validated bedside tool commonly used tocols [29]. for clinical diagnosis of PSOD [16]. We recently assessed Carnaby et al. [30] obtained similar results to Crary et al. the psychometrics of V-VST through a systematic review [26], showing that patients with acute stroke and dysphagia and found that V-VST has good psychometric properties for had a reduction in SSF (0.27 swallows/min) compared to OD in several phenotypes [16]. Regarding diagnostic accu- patients with acute stroke and without dysphagia (0.51 swal- racy, the V-VST had a sensitivity for OD of 93.17%, 81.39% lows/min) at the time of stroke admission. In addition, they specificity, and an inter-rater reliability Kappa of 0.77. Also, showed a reduction in SSF at the time of discharge (0.34 likelihood ratios (LHR) for OD were 0.08 (LHR–) and 5.01 vs 0.47 swallows/min) and an association between reduced (LHR +), and the diagnostic odds ratio for OD was 51.18 SSF and greater severity of dysphagia and stroke, evaluated [16]. Specifically for clinical signs of safety impairment, the in terms of disability and institutionalization. V-VST presented a sensitivity of 88.2% and specificity of Recently, our group demonstrated a decline in SSF with 64.7%; for pharyngeal residue, it had a sensitivity of 88.4% aging and an increase following peripheral oropharyngeal and specificity at 34.6% [17]. sensory stimulation [31]. The mean of SSF in young healthy Normal swallowing requires the participation of a volunteers was around one swallow per minute and was sig- widely extended cerebral network with final output motor nificantly reduced to half by older age but not by gender with commands occurring after activation of brainstem circuits a significant negative correlation between age and SSF [31]. involving the nuclei of the central pattern generators [18, Little is known yet about reflexive swallow mechanisms and 19]. The integrity of this complex top-to-bottom neural axis the role of spontaneous swallowing frequency in OD. As is necessary for the coordination of the oropharyngeal swal- Bulmer et al. [32] found, there is a negative association of low response during volitional swallows and ideal for reflex- SSF with age, which could be implicated in dysphagia. How- ive or spontaneous swallows in order to adapt the motor ever, they conclude in their review that although dysphagia response to the passage of the oral content through the oro- has been associated with lower SSF there is still high het- pharynx [20, 21]. Spontaneous swallowing is, therefore, a erogeneity in SSF data and methodologies used to evaluate protective aerodigestive oligosynaptic brainstem reflex, the it, that it is difficult to assess the cause and effect of SSF primary trigger of which depends on a “peripheral” stimulus on dysphagia and that age and comorbidities are potential (saliva) but could also involve a central source and modula- confounding factors [32]. tion mechanism [21] although this aspect is still under dis- Saliva plays a role in swallowing, helping digestion, cussion. Despite evidence that reflexive swallow still occurs and bolus formation. Saliva quantity and quality can be in decerebrated animals [18], in humans this reflex response affected by multiple diseases and medical treatments may be modulated by cortical descending influences similar [33]. In normal conditions daily saliva production ranges to other brainstem reflexes [19, 22, 23]. Spontaneous swal- between 0.5 and 1.5 L. When unstimulated, saliva flow rate lowing (SS) physiologically contributes to the protection of is 0.3–0.4 mL min. Salivary flow is affected by conscious the respiratory tract by clearing saliva and its impairment state, decreasing to 0.1 mL/min when sleeping. In addi- could lead to an increase in pharyngeal secretions and the tion, an increase is observed while eating (4.0–5.0 mL/min) risk of aspiration [24]. [33]. Substance P (SP), is a neuropeptide commonly present Previous studies have shown that spontaneous swallow- in saliva that acts as a neurotransmitter in the pharyngeal ing frequency (SSF) is reduced in patients with dysphagia mucosa in response to local stimuli and is thought to have secondary to multiple medical conditions (hospitalized older an important role in the swallowing reflex, following the patients, stroke, Parkinson's disease, etc.) [24–28]. In the mechanism of action described for the cough reflex [34]. field of stroke, measurement of SSF has been proposed as Previous studies reported that SP concentration in sputum is a potential index for the screening of dysphagia in acute reduced in older and PS patients with aspiration pneumonia stroke patients [26]. In the study of Crary et al., for patients [35, 36] and that a low concentration of SP in saliva would with an acute stroke and OD, SSF was also reduced to half discriminate against those patients with reduced SSF and (0.23 ± 0.15 swallows/min) that of the subgroup without OD pharyngeal sensitivity [25, 37]. SP has also been used as a (0.56 ± 0.31 swallows/min); and SSF correlated negatively potential marker of neurorehabilitation peripheral treatments and significantly with stroke and swallow severity indices for OD [38, 39]. but not with age, time from stroke onset, or consciousness Because the neural control and modulation of sponta- level [26]. They found that a SSF lower than 0.40 could neous swallowing and the mechanisms involved are still 13 M. Alvarez-Larruy et al.: Post-stroke spontaneous swallowing frequency and dysphagia largely unknown in humans, in the present study we aimed blood and saliva were collected, the nutritional status was to compare SSF, salivary SP concentration, hydration and screened with the Mini Nutritional Assessment short form nutritional status of PS patients with and without oropharyn- (MNA-sf) [41], and body composition was assessed by geal dysphagia. We hypothesized that PSOD patients will bioimpedance. In addition, all PS patients were screened present reduced SSF and SP levels, and poor nutritional and using V-VST in order to assess the clinical signs of impaired hydration status when compared to PS patients without OD safety and efficacy of swallow and to classify them into hav- (PSnOD). ing dysphagia (PSOD) or not having dysphagia (PSnOD). In the afternoon, the SSF was recorded for 10 min and an additional saliva sample was collected. In addition, demo- Materials and Methods graphic characteristics, medical history, usual medication, functional status, and stroke severity scales were also col- Study Population lected on admission. Baseline functional status was defined according to the modified Rankin scale (mRS) [42] and the Forty-five PS patients were recruited during their admis- Barthel index [43]. Stroke characteristics included the time sion to the Neurology Department at the spital de Mataró, between stroke onset and assistance, National institute of Barcelona, Spain. The inclusion criteria were to be 18 years Health Stroke Scale (NIHSS) [44], current and previous old or older, to have hemorrhagic or ischemic unilateral neuroimaging lesions, Oxford stroke classification [45], stroke diagnosed by a neurologist using clinical and imaging lateralization, affected arterial territory, and ischemic or tests—computed tomography or magnetic resonance imag- hemorrhagic stroke etiology. ing—no more than 15 days from stroke onset. Exclusion criteria were to have a life expectancy of less than 3 months, severe dementia (Global Deterioration Scale [40] greater Study Variables and Procedures than 5) or suspected or confirmed SARS-COV-2 infection (by RT-PCR 48 h before the visit). All included patients Volume‑Viscosity Swallowing Test (V‑VST) tested negative for Sars-CoV-2 on admission. All patients were assessed with the V-VST [17]. The swal- Study Design lowing function was assessed while swallowing boluses of 5, 10, or 20 mL of < 50 mPa·s (liquid), 250 mPa·s (nectar), The study was an observational transversal study conducted or 800 mPa·s (honey). 250 and 800 mPa·s were obtained by with the cooperation of researchers from the Gastrointesti- adding 2.0 g and 5.5 g of Nutilis Clear (Nutricia, Barcelona, nal Physiology Lab and the Neurology Department of Hos- Spain), respectively, to 100 mL of water. In order to analyze pital de Mataró. The study protocol was approved by the the effect of oropharyngeal dysphagia on the variables, the Ethical Committee of the Hospital de Mataró (CEIm 74/20) patients were divided in two main groups after the V-VST: and performed according to the rules of the Declaration of post-stroke patients with dysphagia in one group (PSOD) Helsinki. All the participants signed the informed consent. and without dysphagia in the other (PSnOD). Patients who Patients who agreed to participate had two visits performed presented any sign of safety and/or efficacy impairments, one in the morning and the other in the afternoon of the except for piecemeal deglutition with 20 ml at 800 mPa.s same day (Fig. 1). During the morning visit a sample of (honey viscosity), were included in the PSOD group. A Fig. 1  Study design. BIA bioimpedance, V-VST volume viscosity swallow test, MNA-SF mini nutritional assessment- short form 13 M. Alvarez-Larruy et al.: Post-stroke spontaneous swallowing frequency and dysphagia subgroup was also considered with patients with clinical Germany) was used to collect the saliva, putting the swab signs of impaired safety of swallow according the V-VST. under the tongue of the patient for 5 min. After the cen- trifugation (at 2600 rpm for 2 min) the clean sample was Neurophysiological Assessment of Spontaneous Swallow stored at −80 °C until the determination of SP using specific (SS) commercial enzyme-linked immunosorbent assay (ELISA) kits: Substance P Parameter Assay Kit (ref. KGE007; R&D The neurophysiological recording method for volitional systems, Minneapolis, MN, USA). swallow assessment previously described by Ertekin et al. [46, 47] was used to characterize SS according to kinematic Hydration Status. Body Composition measures. Surface electromyography (EMG) and acceler- ometry were used to record suprahyoid complex and laryn- Body composition was measured using bioimpedance [37, geal activation, respectively, as previously described by our 48] with the InBody S10 (InBody Co., Ltd., Seoul, Korea), team [31]. A pair of 2 × 2 cm adhesive EMG electrodes were by placing an electrode on the thumb and ring finger of positioned over the suprahyoid muscles, while a 1.5 × 1.5 cm each hand and one more pair on each ankle with the patient omnidirectional accelerometer (Pro-Tech Limb Movement in a lying or sitting position depending on their condition. Sensors-PLM, RespironicsⓇ Inc., Nevada, USA) was placed The parameters collected were the percentage of total body over the cricothyroid cartilage. Online recordings from both water, extracellular water, intracellular water, fat mass and the EMG signals (bandpass 100–5000 Hz) and accelerom- muscle mass and the phase angle (in degrees). The reference eter activity (bandpass < 0.001–10 Hz) were entered into a ranges for total body water, extracellular water, intracellular polygraphic recording system (Biopac Systems Inc., Califor- water and fat and muscle mass were automatically calcu- nia, USA). An additional online calculation channel display- lated by the equipment according to the age, sex, height, ing the rectified and integrated activity from the EMG chan- and weight of each participant. Phase angle normality level nel was built. All recordings were analyzed offline using the was established as equal or greater than 4.40º according to AcqKnowledge software (Biopac, USA), which displayed our previous studies [37]. Reduction of the phase angle is a visual trace of all three channels. Spontaneous swallows related to cell dehydration that compromises the integrity were considered to be present when the following two crite- of the plasmatic membrane, altering its capacity to separate ria were fulfilled: (1) the activity recorded on the EMG and charges of the bioimpedance electric current [37]. Phase the accelerometer’s channels was compatible with a typical angle is also linked to the quality of the muscle function in neurophysiological swallow configuration as described [46], older adults [49, 50]. and (2) the previous recording activity was consistent with SS through visual inspection by the investigator. A base- line 10 min recording was performed in each patient for SS Nutritional Status. Mini Nutritional Assessment Short Form characterization. (MNA‑SF) and Blood Analysis Neurophysiological Data Reduction and Analysis After the In order to screen for risk of malnourishment, the patients 10’-recording, we obtained the following data for each indi- completed the MNA-sf [41]. If a patient had difficulty with vidual: total number of SS and number of swallows per min- comprehension/expression, a family member or caregiver ute. Our primary physiological outcome was the SSF, which was interviewed. In addition, blood samples were collected was calculated as the total number of SS during the record- as part of routine care during the patient admission. From ing divided by 10 (expressed as swallows/min). Addition- the blood analysis, we collected data on the values of albu- ally, in order to characterize SS performance of study popu- min, prealbumin, protein, lymphocytes, and total choles- lations, the following kinematic measures were calculated in terol. Blood analytical parameter reference intervals were patients for each SS [31]: amplitude (distance from highest taken from the Reference Laboratory of Catalonia [51]: for to lowest peak of the EMG signal, expressed in volts), dura- albumin they were 3.5 to 5.2 g/dL; for prealbumin, 20.0 to tion, delta T (∆T, time from the start of the EMG signal to 40.0 g/dL; total protein level was considered normal between the end in seconds), and area under the curve (AUC, ampli- 6.0 and 8.3 g/dL; lymphocytes, between 1 × ­103 and 3 × ­103/ tude and ∆T integral, in volts·seconds). µL; and total cholesterol, 120.0 to 200.0 mg/dL. Saliva Sample Collection and SP Quantification Functional Outcome Scale Score (FOIS) [52] Saliva samples were collected in the morning (from 8 am A seven-point ordinal scale was used to document the func- to 12 noon) and also in the afternoon (from 1:30 to 5 pm) tional level of oral intake of food and liquid based on the diet from all participants. A Salivette® (Sarstedt, Nuembrecht, prescriptions from the assessment day. 13 M. Alvarez-Larruy et al.: Post-stroke spontaneous swallowing frequency and dysphagia Power Analysis in sex distribution between groups. The post-stroke level of independence for daily activities (Barthel Index) decreased We conducted a sample size analysis before starting data in both groups with significant differences (p = 0.0003). In collection: Accepting an alpha risk of 0.05 and a beta risk addition, the severity of stroke was significantly higher in of 0.2 in a two-sided test, a minimum of 20 persons are PSOD, demonstrated by moderate stroke severity against necessary in the first group and 20 in the second to find minor stroke in the PSnOD group (p =  < 0.0001). a statistically significant difference in proportion, expected to be 0.35 in the OD group and 0.8 in the control group. A Volume‑Viscosity Swallowing Test (V‑VST) drop-out rate of 15% was anticipated. This was calculated on the expected difference in SSF between patients with and Post‑Stroke Patients Without Dysphagia (PSnOD) without OD. Eighteen patients did not present impaired efficacy or safety Data Analysis and Statistical Methods of swallowing during the V-VST. Continuous data were expressed as mean ± standard devia- Post‑Stroke Patients with Dysphagia (PSOD) tion (SD) and categorical data as absolute or relative fre- quencies. Continuous data were analyzed by unpaired t-test Twenty-seven patients (60%) presented clinical signs of OD (PSOD vs PSnOD, PSOD with impaired safety vs PSnOD) according to the V-VST. Impaired efficacy and impaired or pair t-test (pre-admission vs discharge) while categorical safety was observed in 42.20%. Eight (29.63%) patients data were compared by Fisher’s exact test. A non-parametric presented only efficacy impairment and 19 (70.37%) had test was performed for non-Gaussian data. Receiver-operator safety impairment. From the group with safety impair- characteristic (ROC) curves were drawn to explore the cut- ment, 6 (31.58%) presented only safety impairment and 13 off SSF to assess the accuracy of the clinical diagnosis of (68.42%) both safety and efficacy impairment. dysphagia and impaired safety of swallow. Statistically sig- Regarding impaired safety, voice changes were present nificant differences were considered when p-value was less in 8.89% with nectar and 7.89% with liquid (p = 0.1371). than 0.05 and all the analyses were performed using Graph- Oxygen desaturation was observed in 2.22% with nectar and Pad Prism 6 (GraphPad Software, San Diego, CA, USA). honey (p = 0.6512). Cough was seen in 6.67% for nectar vis- cosity and 42.11% with liquid (nectar vs liquid p < 0.0001; honey vs liquid p < 0.0001), demonstrating a therapeutic Results effect of this thickening agent improving swallow safety (Fig. 2). Demographic and Stroke Characteristics For impaired efficacy, lip seal impairment was observed in 6.67% with nectar and 4.44% with honey viscosity Forty-five post-stroke patients, 28 men (62.22%), with a (p = 0.2875). Oral residue reached 6.67% with nectar and mean age of 71.78 (13.46) years old were included in this 11.11% with honey (p = 0.1130). Piecemeal deglutition was study. Mean time between the stroke onset and inclusion in seen in 37.78% with nectar and honey, and 10.53% with liq- the study was 4.98 (2.80) days. Severity of strokes according uid (nectar vs liquid p = 0.0053; honey vs liquid p = 0.0053; to NIHSS was mild (Table 1) (Table SI). nectar vs honey p = 1.000). Pharyngeal residue was detected Most patients had ischemic (84.44%) strokes and also in 2.22% with nectar and 11.11% with honey (p = 0.0362; most of them were supratentorial (86.66%) strokes affect- nectar vs liquid p = 1.000; honey vs liquid p = 0.2599; nectar ing purely the anterior circulation brain territory (81.57%). vs honey p = 0.2028). Proportion of left and right hemisphere involvement was the same. The majority of patients (46.67%) also showed Spontaneous Swallow Characterization in PSnOD leukoaraiosis. The etiology of ischemic stroke was athero- and PSOD Patients. SSF in Patients with PSOD thrombotic in 26.32% of patients, cardioembolic in 23.68%, and Impaired Safety of Swallow lacunar in 15.79%, and indeterminate in 34.21%. All hemor- rhagic strokes (n = 7; 15.56%) were of hypertensive etiology. SSF was 0.48 ± 0.29 swallows per minute in PSnOD patients, and was significantly reduced to half (0.23 ± 0.18 PSOD versus PSnOD Demographic and Clinical swallows per minute) in PSOD patients (N = 27, p = 0.0017). Characteristics In PSnOD, the mean amplitude was 0.64 ± 1.62 V, the mean delta T 1.66 ± 0.27 s, and the mean AUC PSOD patients were significantly older than PSnOD patients 0.07 ± 0.07 V·seconds. For PSOD, the mean amplitude was (p = 0.0015), (Table 1). However, there were no differences 0.76 ± 1.25 V, the mean delta T was 1.81 ± 0.90 s, and the 13 M. Alvarez-Larruy et al.: Post-stroke spontaneous swallowing frequency and dysphagia Table 1  Demographic and All PSOD PSnOD p-value neuroimaging characteristics of the study population N 45 27 18 Sociodemographic Data Sex (% men) 62.22 (28) 55.56 (15) 72.22 (13) 0.3513 Age (years) 71.78 ± 13.46 76.81 ± 12.34 64.22 ± 11.37 0.0015 Barthel Index pre-admission (mean ± SD) 95.56 ± 10.71 93.33 ± 13.08 98.89 ± 4.71 0.0818 Barthel Index admission (mean ± SD) 60.56 ± 32.01 47.48 ± 31.97 79.72 ± 22.39 0.0003 MNA-sf (mean ± SD) 10.82 ± 1.74 10.22 ± 1.93 11.72 ± 0.80 0.0009 NIHSS (mean ± SD) 5.09 ± 5.62 7.52 ± 6.17 1.44 ± 1.34 < 0.0001 Stroke characteristics Days from ictus to evaluation (mean ± SD) 4.98 ± 2.80 4.39 ± 2.48 5.37 ± 3.03 0.2326 mRS pre-admission (%) 0–2 80.00 70.37 94.44 0.0644 3–5 20.00 29.63 5.56 0.0644 Stroke subtype (%) Ischemic 84.44 81.48 88.89 0.6844 Hemorrhagic 15.56 18.52 11.11 0.6844 Acute lesions in neuroimaging (%) No lesions 4.44 0 11.11 0.1545 Territorial infarction 68.89 77.78 72.22 1.000 Lacunar infarction 11.11 11.11 11.11 1.000 Intraparenchymal hemorrhage 15.56 18.52 11.11 0.6844 Lateralization (%) Right hemisphere 42.22 44.44 38.89 0.7660 Left hemisphere 44.44 44.44 44.44 1.000 Brain stem 13.33 11.11 16.67 0.6703 Oxford (%) PACI 24.44 18.52 33.33 0.3040 TACI 24.44 37.04 5.56 0.0307 POCI 6.67 3.70 16.67 0.2862 LACI 42.22 40.74 44.44 1.000 PACI partial anterior circulation infarcts, TACI total anterior circulation infarcts, POCI posterior circulation infarcts, LACI lacunar infarcts mean AUC 0.14 ± 0.22 V-second. Considering PSOD with the best cut-off value (Fig. 3). Considering the SSF, the cut- clinical signs of impairment in swallow safety (n = 19), we off value to predict dysphagia in PSOD was 0.55 swallows/ observed that SSF was 0.22 ± 0.18 swallows per minute, minute, with a sensitivity of 0.92, a specificity of 0.33, and and also significantly decreased by half in comparison with AUC of 0.76 (95% CI: 0.62–0.90). However, when consid- PSnOD (p = 0.0024). For PSOD with safety impairment, ering PSOD with safety impairment, the cut-off value to the mean amplitude was 0.42 ± 0.50 V, the mean delta T predict unsafe swallows was 0.45 swallows/minute, with a was 1.63 ± 0.40 s, and the mean AUC 0.07 ± 0.05 V-second. sensitivity of 0.95, a specificity of 0.44, and AUC of 0.76 No significant differences on these metrics were observed (95% CI: 0.63–0.90). when comparing PSOD or PSOD with safety impairments vs PSnOD (p > 0.05). SP Levels in Saliva ROC Curve to Predict Oropharyngeal Dysphagia and Impaired Swallow Safety In the saliva collected in the morning, SP concentra- tion was 62.02 ± 26.23 pg/ml in the PSnOD group and ROC curve analysis of SSF was developed to assess the 60.55 ± 26.08 pg/ml for PSOD. In the afternoon, the con- accuracy of the clinical diagnosis of dysphagia and impaired centration for PsnOD patients was 76.28 ± 57.88 pg/ml and safety of swallow used to predict dysphagia and to select 78.80 ± 63.19 pg/ml in the PSOD group. No differences were 13 M. Alvarez-Larruy et al.: Post-stroke spontaneous swallowing frequency and dysphagia Fig. 2  Effect of fluid thickening at increased shear viscosity val- ues on safety (250 mPa.s vs thin liquid p < 0.0001; 800 mPa.s vs thin liquid p < 0.0001) of swal- low in PSOD. Note the strong therapeutic effect of increasing bolus viscosity on safety of swallow Fig. 3  SSF receiver-operator characteristic (ROC) curve show- B. Depicted is the cut-off value of 0.45 swallows per minute for ing spontaneous swallowing frequency sensitivity/ specificity for impaired safety of swallow. PSOD post-stroke patients with oro- oropharyngeal dysphagia (A) or impaired safety of swallow (B). A. pharyngeal dysphagia, SSF spontaneous swallowing frequency Depicted is the cut-off value of 0.55 swallows per minute for OD. found between groups for both morning (p = 0.7524) and malnourishment was observed in the PSOD group (PSOD afternoon (p = 0.4121) SP concentration in saliva. 48.15% vs PSnOD 11.11%; p = 0.0118). There were no dif- ferences regarding the prevalence of malnutrition for both Nutritional Status. MNA‑SF and Blood Biomarkers groups (PSOD 14.81% vs PSnOD 0%; p = 0.1383). Blood biomarkers showed lower mean albumin (p = 0.0224), The mean MNA-sf score was significantly reduced in prealbumin (p = 0.0128), and protein (p = 0.0419) in PSOD (10.22 ± 1.97) when compared to the PSnOD PSOD (Table 2). There were no differences for lympho- group (11.72 ± 0.83) (p = 0.0009). Normal nutritional cytes (p = 0.3142) and total cholesterol (p = 0.1483) lev- status was observed in 37.04% of PSOD vs 88.89% of els. These results show hypermetabolism and changes in PSnOD (p = 0.007). Also, high prevalence of risk of 13 M. Alvarez-Larruy et al.: Post-stroke spontaneous swallowing frequency and dysphagia Table 2  Serum nutritional biomarkers in PSOD versus PSnOD (p < 0.005 vs PSnOD), showing that the majority of them PSOD PSnOD p-value needed a texture modified diet level E (fork mashable) or C (puree like) (N = 18); 10 patients (55.56%) needed nectar Albumin (g/dL) 3.75 ± 0.53 4.28 ± 0.93 0.0224 (250 mPa·s); 8 patients (44.44%) needed honey (800 mPa·s) Prealbumin (g/dL) 17.99 ± 5.87 22.71 ± 5.10 0.0128 viscosity to maintain a safe swallow while swallowing fluids Total Proteins (g/dL) 6.61 ± 0.46 6.88 ± 0.37 0.0419 and 9 patients did not need thickened fluids. Lymphocytes (µL) 1.97 ± 1.07 2.18 ± 0.84 0.3142 Total cholesterol (mg/dL) 147.29 ± 38.81 162.14 ± 38.52 0.1483 PSOD post-stroke patients with oropharyngeal dysphagia, PSnOD Discussion post-stroke patients without oropharyngeal dysphagia The main objective of this study was to assess whether SSF is decreased in patients with an acute stroke and dysphagia. Table 3  Body composition in PSOD versus PSnOD The secondary objectives were to analyze whether there is PSOD PSnOD p-value a decrease in the levels of substance P in saliva and a worse nutritional and hydration status in patients with an acute Total water (%) 35.13 ± 9.49 39.00 ± 7.01 0.0664 stroke and dysphagia. The prevalence of OD among acute Intracellular water (%) 21.47 ± 5.87 24.23 ± 4.57 0.0570 stroke patients was 60.00%, higher than in our previous Extracellular water (%) 13.66 ± 3.64 14.77 ± 2.48 0.0997 studies [7]. This might be attributed to 1) the higher sever- Muscle mass (%) 26.00 ± 7.65 29.59 ± 5.95 0.0552 ity of stroke events in this study than in the previous one Fat mass (%) 30.03 ± 14.04 27.91 ± 12.87 0.7595 (NIHSS mean of 5 compared to 2) possibly due to the fact Phase angle (degrees) 4.96 ± 1.03 6.01 ± 1.15 0.0054 that patients with mild stroke were not admitted to neurology PSOD post-stroke patients with oropharyngeal dysphagia, PSnOD ward during COVID-19 pandemia, and 3) that all patients post-stroke patients without oropharyngeal dysphagia in this study were evaluated by the same experienced pro- fessional (WN). Regarding the type, topology, and etiology of the stroke, we obtained very similar characteristics to nutritional status. Also, total protein, albumin, and preal- those described in our previous studies and in the literature bumin are good markers for acute changes in nutritional [7]. Both impaired efficacy and impaired safety of swallow status. were observed in 42.20% of the PSOD patients in the current study. The main sign of impaired safety was cough and the Hydration Status. Bioimpedance main sign of impaired efficacy was piecemeal deglutition. The main result of this study is that SSF was markedly Bioimpedance measurements showed that there was no reduced in PSOD patients compared to PSnOD, (0.23 ± 0.18 alteration in the mean angle phase in PSnOD group; how- swallows/minute) compared to PSnOD (0.48 ± 0.29 swal- ever, 37.50% of PSOD presented impairment suggesting lows/minute). These results are very similar to those intracellular dehydration (p = 0.0054). (Table 3). Regarding obtained by Crary et al. [26], who also found a reduction bioimpedance nutritional measurements/body composition, to half of SSF (0.23 ± 0.15 swallows/min in PSOD patients we observed a tendency, not statistically significant, to lower vs 0.56 ± 0.31 swallows/min in PSnOD patients), and those percentage levels of total water, intracellular water and mus- obtained by Carnaby et al. [30] (0.27 ± 0.22 swallows/min cle mass in the PSOD group. There was no difference for vs 0.51 ± 0.30 swallows/min). In addition, Crary et al. [29] extracellular water and fat mass between groups. suggested that SSF may be a useful screening tool, but it lacked the specificity for a definitive diagnosis. Similarly to FOIS Crary et al.’s [26] study, we made a ROC curve to assess the accuracy of measuring SSF as a screening tool for dysphagia All patients included in the study were fed exclusively by and impaired safety of swallow, using the cut-off values of mouth but there was a significant difference between swal- 0.55 swallows/minute for PSOD and 0.45 swallows/minute lowing functionality between groups (< 0.0001) showing for PSOD with impaired safety. The results showed that SSF that PSOD patients were in a need for modified texture diet has quite good diagnostic accuracy, and a reduction in SSF and fluid thickening, and this was not an issue for the PSnOD presents high sensitivity, but low specificity in PSOD. group. All PSnOD patients were eating a normal texture In previous studies we also found a significant decrease diet with thin liquid for hydration [53] except for one, who in pharyngeal sensory function associated with aging, much needed an level E diet (easy to chew) because of tooth loss; more severe in older patients with dysphagia [54]. In patients they presented a mean FOIS score of 6.89 ± 0.46. The mean with PSOD and impaired safety of swallow, videofluoro- FOIS score for the PSOD group however was 4.70 ± 0.81 scopic studies identified delayed laryngeal vestibule closure 13 M. Alvarez-Larruy et al.: Post-stroke spontaneous swallowing frequency and dysphagia time as the critical event causing penetrations and aspira- to note that there was only a small number of patients who tions, and our neurophysiological evaluation demonstrated had enough saliva to be analyzed using ELISA affecting, as prevalent impairments with disrupted integration of both a consequence, the statistical power of the salivary neuro- pharyngeal sensory inputs and reduced cortical excitability peptides results. of efferent pathways [55]. For the future, the psychometrics of SSF as a screening The reduction in SSF we thought could be due to the method for PSOD show very good sensitivity and limited decrease in the pharyngeal sensory stimulus to the central specificity that suggests the need for its combination with pattern generator secondary to the stroke. Reduced SSF more specific clinical and/or instrumental methods for could lead to an increase in the accumulation of oropharyn- PSOD. In addition, and taken together with our previous geal secretions and, therefore, to an increased risk of aspira- studies on SSF and its potential relation with impaired phar- tion and pneumonia [24]. We also found in previous studies yngeal sensory function, we believe SSF might become a that increasing pharyngeal sensory input in post-stroke dys- method to assess the effect of pharyngeal pharmacologic phagic patients by using TRPV1 (transient receptor potential stimulation on the swallow response in PSOD. cation channel subfamily V member 1) stimulants almost doubled SSF, further supporting the role of pharyngeal sen- sory stimulation in the modulation of SSF and the potential Summary use of SSF as a relevant metric in assessing the pharmaco- logical stimulation of swallow response in PSOD [31]. In summary, SSF is significantly reduced in PSOD, and has a On the other hand there wasn’t a significant difference for high prevalence of clinical signs of impaired safety of swal- SP levels in saliva in PSOD vs PSnOD. Nutritional status low. Our study further confirms that acute PSOD patients was poor in up to 60.00% of PSOD patients and dehydration present hydropenia, poor nutritional status, and high risk for was observed in 37.50% of them. These results further show respiratory complications. PSOD patients are at great risk of malnutrition, dehydration, and respiratory complications. Supplementary Information The online version contains supplemen- One of the open questions about the study of the SSF tary material available at https://d​ oi.o​ rg/1​ 0.1​ 007/s​ 00455-0​ 22-1​ 0451-3. is what is the most appropriate method to evaluate it [32]. Our study not only demonstrates the reproducibility of the Acknowledgements We acknowledge the nurses and the nursing assis- tants from the Neurology Department at the Hospital de Marató, the previous studies [26, 29, 30], but also the reproducibility by statistical help received from Elisabet Palomera and we thank Jane means of another SSF evaluation technique. While previous Lewis for English editing assistance. studies [26, 29, 30] measured SSF with an acoustic record- ing obtained through a miniature microphone, in our study Author’s contribution Data acquisition: MA, NT, NG, WN; Data anal- SSF was measured with EMG. More studies are required ysis: NT, WN; Data interpretation: NT, PC, WN; Conceptualization and work design: PC, WN; Writing: MA, NT, WN; Revision: EP, PC, to analyze whether SSF could be a screening indicator for WN; Final approval: PC, WN. dysphagia in acute stroke in combination with more specific clinical and/or instrumental assessments. Funding Open Access Funding provided by Universitat Autonoma Our study presents some limitations. We included a mod- de Barcelona. This study was supported by research grants from: erate sample size of 45 post-stroke patients which possibly Investigación Clínica Independiente, Instituto de Salud Carlos III (ICI20/00117); CIBERehd, Acción Estratégica en Disfagia, Instituto limited the interpretation of the statistical results. A larger de Salud Carlos III (EHD20PI02); Fondo de Investigaciones Sani- sample size is necessary to study this topic in greater depth. tarias, Instituto de Salud Carlos III (PI18/00241); and The Territorial In addition, carrying out the study during COVID-19 pan- Competitiveness Specialization Project (PECT) of Mataro-Maresme demia could have generated a bias toward the selection of (PRE/161/2019) financed by the Government of Catalunya-Generalitat de Catalunya within the framework of the European Regional Develop- patients with more severe strokes than in our previous stud- ments Funds of Catalonia Operational Programme 2014–2020. ies. Furthermore, in this study dysphagia was determined by V-VST which, despite being a test with excellent psy- Declarations chometric properties for clinical diagnosis of OD, is not the gold standard test for OD, which is the videofluoroscopy. In Conflict of interests The authors have no conflicts of interest to declare addition, although the V-VST uses some clinical symptoms, that are relevant to the content of this article. it also uses several signs to identify impaired safety and effi- cacy of swallow. Finally, although the role of SP in swallow- Open Access This article is licensed under a Creative Commons Attri- bution 4.0 International License, which permits use, sharing, adapta- ing is still unknown, there are some studies describing its tion, distribution and reproduction in any medium or format, as long relationship with several complications associated with OD, as you give appropriate credit to the original author(s) and the source, such as aspiration pneumonia, decreased SSF, and impaired provide a link to the Creative Commons licence, and indicate if changes pharyngeal sensitivity [25], [35–37]. Finally, it is important were made. The images or other third party material in this article are 13 M. Alvarez-Larruy et al.: Post-stroke spontaneous swallowing frequency and dysphagia included in the article's Creative Commons licence, unless indicated swallow test for clinical evaluation of oropharyngeal dysphagia. otherwise in a credit line to the material. If material is not included in Neurogastroenterol Motil. 2014;26:1256–65. https://​doi.​org/​10.​ the article's Creative Commons licence and your intended use is not 1111/​nmo.​12382. permitted by statutory regulation or exceeds the permitted use, you will 16. Riera SA, Marin S, Serra-Prat M, et al. A systematic and a scoping need to obtain permission directly from the copyright holder. To view a review on the psychometrics and clinical utility of the volume-vis- copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. cosity swallow test (V-VST) in the clinical screening and assess- ment of oropharyngeal dysphagia. Foods. 2021;10:1900. https://​ doi.​org/​10.​3390/​foods​10081​900. 17. Clavé P, Arreola V, Romea M, et al. Accuracy of the volume- References viscosity swallow test for clinical screening of oropharyngeal dys- phagia and aspiration. Clin Nutr. 2008;27:806–15. https://d​ oi.o​ rg/​ 10.​1016/j.​clnu.​2008.​06.​011. 1. Avan A, Digaleh H, Di Napoli M, et al. Socioeconomic status 18. Jean A. Brain stem control of swallowing: neuronal network and and stroke incidence, prevalence, mortality, and worldwide cellular mechanisms. Physiol Rev. 2001;81:929–69. https://​doi.​ burden: an ecological analysis from the Global Burden of Dis- org/​10.​1152/​physr​ev.​2001.​81.2.​929. ease Study 2017. BMC Med. 2019. https://​doi.​org/​10.​1186/​ 19. Michou E, Hamdy S. Cortical input in control of swallowing. s12916-​019-​1397-3. Curr Opin Otolaryngol Head Neck Surg. 2009;17:166–71. 2. Wafa HA, Wolfe CDA, Emmett E, et al. Burden of stroke in https://​doi.​org/​10.​1097/​MOO.​0b013​e3283​2b255e. Europe. Stroke. 2020;51:2418–27. https:// ​ d oi. ​ o rg/ ​ 1 0. ​ 1 161/​ 20. Hamdy S, Rothwell JC, Brooks DJ, et al. Identification of the STROK​EAHA.​120.​029606. cerebral loci processing human swallowing with H 2 15 O PET 3. Luengo-Fernandez R, Violato M, Candio P, Leal J. Economic activation. J Neurophysiol. 1999;81:1917–26. https://​doi.​org/​ burden of stroke across Europe: a population-based cost analysis. 10.​1152/​jn.​1999.​81.4.​1917. Eur Stroke J. 2020;5:17–25. https://​doi.​org/​10.​1177/​23969​87319​ 21. Kamarunas E, Mulheren R, Palmore K, Ludlow C. Tim- 883160. ing of cortical activation during spontaneous swallowing. 4. Marin S, Serra-Prat M, Ortega O, et al. Healthcare costs of post- Exp Brain Res. 2018;236:475–84. https://​d oi.​o rg/​1 0.​1 007/​ stroke oropharyngeal dysphagia and its complications: malnutri- s00221-​017-​5139-5. tion and respiratory infections. Eur J Neurol. 2021;28(11):3670– 22. Berardelli A, Accornero N, Cruccu G, et al. The orbicularis oculi 81. https://​doi.​org/​10.​1111/​ene.​14998. response after hemispheral damage. J Neurol Neurosurg Psych. 5. Martino R, Foley N, Bhogal S, et al. Dysphagia after stroke. 1983;46:837–43. https://​doi.​org/​10.​1136/​jnnp.​46.9.​837. Stroke. 2005;36:2756–63. https://​doi.​org/​10.​1161/​01.​STR.​00001​ 23. Cabib C, Llufriu S, Martinez-Heras E, et al. Abnormal control 90056.​76543.​eb. of orbicularis oculi reflex excitability in multiple sclerosis. PLoS 6. Heuschmann PU, Wiedmann S, Wellwood I, et al. Three-month One. 2014;9:e103897. https://d​ oi.o​ rg/1​ 0.1​ 371/j​ ourna​ l.p​ one.0​ 1038​ stroke outcome: the European Registers of Stroke (EROS) inves- 97. tigators. Neurology. 2011;76:159–65. https://​doi.​org/​10.​1212/​ 24. Murray J, Langmore SE, Ginsberg S, Dostie A. The significance WNL.​0b013​e3182​06ca1e. of accumulated oropharyngeal secretions and swallowing fre- 7. Rofes L, Muriana D, Palomeras E, et al. Prevalence, risk factors quency in predicting aspiration. Dysphagia. 1996;11:99–103. and complications of oropharyngeal dysphagia in stroke patients: https://​doi.​org/​10.​1007/​BF004​17898. a cohort study. Neurogastroenterol Motil. 2018. https://​doi.​org/​ 25. Niimi M, Hashimoto G, Hara T, et al. Relationship between fre- 10.​1111/​nmo.​13338. quency of spontaneous swallowing and salivary substance P level 8. Smithard DG, Smeeton NC, Wolfe CDA. Long-term outcome in patients with acute stroke. Dysphagia. 2018;33:414–8. https://​ after stroke: does dysphagia matter? Age Ageing. 2007;36:90–4. doi.​org/​10.​1007/​s00455-​017-​9867-2. https://​doi.​org/​10.​1093/​ageing/​afl149. 26. Crary MA, Carnaby GD, Sia I, et al. Spontaneous swallowing 9. Finestone HM, Greene-Finestone LS, Wilson ES, Teasell RW. frequency has potential to identify dysphagia in acute stroke. Prolonged length of stay and reduced functional improvement rate Stroke. 2013;44:3452–7. https://​doi.​org/​10.​1161/​STROK​EAHA.​ in malnourished stroke rehabilitation patients. Arch Phys Med 113.​003048. Rehabil. 1996;77:340–5. https://​doi.​org/​10.​1016/​s0003-​9993(96)​ 27. Tanaka N, Nohara K, Kotani Y, et al. Swallowing frequency in 90081-7. elderly people during daily life. J Oral Rehabil. 2013;40:744–50. 10. Crary MA, Humphrey JL, Carnaby-Mann G, et al. Dysphagia, https://​doi.​org/​10.​1111/​joor.​12085. nutrition, and hydration in ischemic stroke patients at admission 28. South AR, Somers SM, Jog MS. Gum chewing improves swallow and discharge from acute care. Dysphagia. 2013;28:69–76. https://​ frequency and latency in Parkinson patients: a preliminary study. doi.​org/​10.​1007/​s00455-​012-​9414-0. Neurology. 2010;74:1198–202. https://​doi.​org/​10.​1212/​WNL.​ 11. Smithard DG, O’Neill PA, Parks C, Morris J. Complications 0b013​e3181​d9002b. and outcome after acute stroke. Does dysphagia matter? Stroke. 29. Crary MA, Carnaby GD, Sia I. Spontaneous swallow frequency 1996;27:1200–4. https://​doi.​org/​10.​1161/​01.​str.​27.7.​1200. compared with clinical screening in the identification of dyspha- 12. Smithard DG, O’Neill PA, England RE, et al. The natural history gia in acute stroke. J Stroke Cerebrovasc Dis. 2014;23:2047–53. of dysphagia following a stroke. Dysphagia. 1997;12:188–93. https://​doi.​org/​10.​1016/j.​jstro​kecer​ebrov​asdis.​2014.​03.​008. https://​doi.​org/​10.​1007/​PL000​09535. 30. Carnaby G, Sia I, Crary M. Associations between spontaneous 13. Hinchey JA, Shephard T, Furie K, et al. Formal dysphagia screen- swallowing frequency at admission, dysphagia, and stroke-related ing protocols prevent pneumonia. Stroke. 2005;36:1972–6. https://​ outcomes in acute care. Arch Phys Med Rehabil. 2019;100:1283– doi.​org/​10.​1161/​01.​STR.​00001​77529.​86868.​8d. 8. https://​doi.​org/​10.​1016/j.​apmr.​2019.​01.​009. 14. Lakshminarayan K, Tsai AW, Tong X, et al. Utility of dyspha- 31. Nascimento W, Tomsen N, Acedo S, et al. Effect of aging, gen- gia screening results in predicting poststroke pneumonia. Stroke. der and sensory stimulation of trpv1 receptors with capsaicin on 2010;41:2849–54. https://​doi.​org/​10.​1161/​STROK​EAHA.​110.​ spontaneous swallowing frequency in patients with oropharyngeal 597039. dysphagia: a proof-of-concept study. Diagnostics. 2021;11:461. 15. Rofes L, Arreola V, Mukherjee R, Clavé P. Sensitivity and speci- https://​doi.​org/​10.​3390/​diagn​ostic​s1103​0461. ficity of the eating assessment tool and the volume-viscosity 13 M. Alvarez-Larruy et al.: Post-stroke spontaneous swallowing frequency and dysphagia 32. Bulmer JM, Ewers C, Drinnan MJ, Ewan VC. Evaluation of spon- Neurophysiol. 2015;32:314–23. https://​doi.​org/​10.​1097/​WNP.​ taneous swallow frequency in healthy people and those with, or 00000​00000​000189. at risk of developing, dysphagia: a review. Gerontol Geriatr Med. 48. Carrión S, Roca M, Costa A, et al. Nutritional status of older 2021;7:1–13. https://​doi.​org/​10.​1177/​23337​21421​10418​01. patients with oropharyngeal dysphagia in a chronic versus an 33. Lorgulescu G. Saliva between normal and pathological. Impor- acute clinical situation. Clin Nutr. 2017;36:1110–6. https://​doi.​ tant factors in determining systemic and oral health. J Med Life. org/​10.​1016/j.​clnu.​2016.​07.​009. 2009;2:303–7. 49. Kilic MK, Kizilarslanoglu MC, Arik G, et al. Association of bio- 34. Canning BJ, Chang AB, Bolser DC, et al. Anatomy and neuro- electrical impedance analysis-derived phase angle and sarcopenia physiology of cough. Chest. 2014;146:1633–48. https://​doi.​org/​ in older adults. Nutr Clin Pract. 2017;32:103–9. https://​doi.​org/​ 10.​1378/​chest.​14-​1481. 10.​1177/​08845​33616​664503. 35. Nakagawa T, Ohrui T, Sekizawa K, Sasaki H. Sputum substance P 50. Yamada M, Kimura Y, Ishiyama D, et al. Phase Angle is in aspiration pneumonia. Lancet. 1995;345:1447. https://​doi.​org/​ a Useful indicator for Muscle Function in Older Adults. J 10.​1016/​S0140-​6736(95)​92638-0. Nutr Health Aging. 2019;23:251–5. https://​doi.​org/​10.​1007/​ 36. Arai T, Yoshimi N, Fujiwara H, Sekizawa K. Serum substance P s12603-​018-​1151-0. concentrations and silent aspiration in elderly patients with stroke. 51. Laboratori de Referencia de Catalunya analytical reference values Neurology. 2003;61:1625–6. https://​doi.​org/​10.​1212/​01.​WNL.​ of the Laboratori de Referencia de Catalunya. n.d. http://​centr​al.​ 00000​96395.​80826.​23. lrc.​cat/​cgi-​bin/​nph-​mgwcgi.​exe?​App¼CATAL​OGO 37. Tomsen N, Ortega O, Nascimento W, et al. Oropharyngeal dys- 52. Crary MA, Mann GDC, Groher ME. Initial psychometric assess- phagia in older people is associated with reduced pharyngeal sen- ment of a functional oral intake scale for dysphagia in stroke sitivity and low substance P and CGRP concentration in saliva. patients. Arch Phys Med Rehabil. 2005;86:1516–20. https://​doi.​ Dysphagia. 2021. https://​doi.​org/​10.​1007/​s00455-​021-​10248-w. org/​10.​1016/j.​apmr.​2004.​11.​049. 38. Suntrup-Krueger S, Bittner S, Recker S, et al. Electrical pharyn- 53. National Patient Safety Agency Dysphagia Expert Reference geal stimulation increases substance P level in saliva. Neurogas- Group (NPSA) (2011) No title. In: Dysphagia diet food texture troenterol Motil. 2016;28:855–60. https://​doi.​org/​10.​1111/​nmo.​ descriptors. http://​www.​hospi​talca​terers.​org/​media/​1160/​dysph​ 12783. agia-​descr​iptors.​pdf 39. van Oosterhout W, Schoonman G, Garrelds I, et al. A human 54. Rofes L, Ortega O, Vilardell N, et al. Spatiotemporal characteris- capsaicin model to quantitatively assess salivary CGRP secre- tics of the pharyngeal event-related potential in healthy subjects tion. Cephalalgia. 2015;35:675–82. https://d​ oi.o​ rg/1​ 0.1​ 177/0​ 3331​ and older patients with oropharyngeal dysfunction. Neurogas- 02414​553824. troenterol Motil. 2017;29: e12916. https://​doi.​org/​10.​1111/​nmo.​ 40. Reisberg B, Ferris S, De Leon M, Crook T. The Global Deteriora- 12916. tion Scale for assessment of primary degenerative dementia. Am J 55. Cabib C, Nascimento W, Rofes L, et al. Neurophysiological and Psychiatry. 1982;139:1136–9. https://​doi.​org/​10.​1176/​ajp.​139.9.​ biomechanical evaluation of the mechanisms which impair safety 1136. of swallow in chronic post-stroke patients. Transl Stroke Res. 41. Kaiser MJ, Bauer JM, Ramsch C, et al. Validation of the mini 2020;11:16–28. https://​doi.​org/​10.​1007/​s12975-​019-​00701-2. nutritional assessment short-form (MNA®-SF): a practical tool for identification of nutritional status. J Nutr Heal Aging. Publisher's Note Springer Nature remains neutral with regard to 2009;13:782–8. https://​doi.​org/​10.​1007/​s12603-​009-​0214-7. jurisdictional claims in published maps and institutional affiliations. 42. Banks JL, Marotta CA. Outcomes validity and reliability of the modified ranking scale: implications for stroke clinical trials. Stroke. 2007;38:1091–6. https://​doi.​org/​10.​1161/​01.​STR.​00002​ 58355.​23810.​c6. Marta Alvarez‑Larruy MD 43. Mahoney FI, Barthel DW. Functional evaluation: the Barthel index. MD State Med J. 1965;14:61–5. Noemí Tomsen PhD 44. Spilker J, Kongable G, Barch C, et al. Using the NIH stroke scale to assess stroke patients. J Neurosci Nurs. 1997;29:384–92. Nicolau Guanyabens MD https://​doi.​org/​10.​1097/​01376​517-​19971​2000-​00008. 45. Bamford J, Sandercock P, Dennis M, et al. Classification and natu- Ernest Palomeras MD, PhD ral history of clinically identifiable subtypes of cerebral infarction. Lancet. 1991;337:1521–6. https://d​ oi.o​ rg/1​ 0.1​ 016/0​ 140-6​ 736(91)​ Pere Clavé MD, PhD 93206-O. 46. Ertekin C, Aydogdu I. Neurophysiology of swallowing. Clin Weslania Nascimento SLP, PhD Neurophysiol. 2003;114:2226–44. https://d​ oi.o​ rg/1​ 0.1​ 016/S ​ 1388-​ 2457(03)​00237-2. 47. Ertekin C. Electrophysiological techniques to evaluate swallow- ing in central and peripheral nervous system disorders. J Clin 13