Introduction

Parkinson’s disease (PD) is a progressive neurodegenerative condition, increasing in prevalence [7]. Oropharyngeal dysphagia and weak cough production are common clinical features in individuals with PD, and aspiration pneumonia is reported to be the leading cause of death in PD [3, 10]. The neurophysiological mechanism underpinning swallow dysfunction is complex and often multifactorial. Many studies have described features of oropharyngeal dysfunction in PD using videofluoroscopic swallow studies and/ or flexible endoscopic evaluation of swallowing (FEES). Tongue pumping and tremor, prolonged bolus transit time, delayed swallow initiation, premature spillage, decreased pharyngeal contraction, reduced hyolaryngeal elevation, uncoordinated airway closure during swallowing, diminished or absent cough response to aspiration, reduced laryngeal sensitivity and post-swallow residue in valleculae and pyriform sinus are described clinical features in PD [1, 5, 17].

In recent years, quantitative changes in swallow pressures have been studied by high-resolution manometry (HRM) [4] which measures important aspects of pharyngeal contractility, upper esophageal sphincter (UES) relaxation, and UES opening. However, literature describing swallow function in PD patients using HRIM is limited. Identifying and describing the pathophysiological swallowing changes in PD will support finding specific interventions that best suit to improve function. Modification of bolus consistencies and volume regulation are commonly used compensatory strategies recommended and employed in dysphagia management, aiming to optimize bolus control and coordination, to facilitate safe swallowing. This study aimed to identify changes in pharyngoesophageal swallowing pressure metrics, in patients with PD, as bolus volume and viscosity were altered. In addition we sought to examine these neuromuscular responses in relation to swallow safety and efficiency using the novel core metrics proposed by an international expert group [15].

Methods

This prospective study was approved by the Health and Disability Ethics Review Committee (HDEC: 19/CEN/131) and written informed consent was obtained from all the participants.

Study population

Consecutive patients with PD with or without complaint of oropharyngeal dysphagia who were referred to our clinic from November 2019 to May 2022 were invited to participate in this study. All the participants were diagnosed with PD by their neurologists and treated with anti-parkinsonism drugs at recruitment. Parkinson’s symptom severity was rated by Parkinson’s Disease Questionnaire-8 (PDQ-8). PDQ-8 is a shortened version of the original PDQ-39. It contains eight questions providing an overall index of self-perceived health in PD [9]. The questions address the following dimensions: mobility, activities of daily living, emotional well-being, social support, cognitions, communication, bodily discomfort, and stigma. Responses are on a 5-point rating scale from ‘never’ to ‘always’ (scores range from 0 to 40 but are then converted to a score out of 100, with 100 representing the greatest severity). No patients had other diagnosed neurological conditions at the time of recruitment. Participants were excluded if they had swallowing complaint related to other diagnosis (e.g. polymyositis, dermatomyositis and muscle dystrophies) or current/ previous head and neck diagnosis and/or treatment (e.g.: head and neck cancer, cervical osteophytes, enlarged tonsils, external compression such as goiter) or were unable/ unwilling to provide informed consent. Participants were all in an ‘on’ state (taking medications) during assessments.

Swallow assessment

Within 24 h of the VFSS and EAT-10 [2], participants also completed a HRIM. Participants were fasting for a minimum period of three hours before the examination and were tested while seated in a head-neutral, upright position for pharyngeal swallow evaluations. A 10-French solid-state unidirectional high-resolution manometry catheter (36 pressure sensors spaced at one cm intervals and 16 adjoining impedance sensors each 2 cm) (Model K103659-E-1180-D, Unisensor AG, Attikon, Switzerland) was inserted transnasally with topical anesthesia (cophenylcaine 4%). Once the catheter was positioned correctly (by viewing manometry readings), participants rested for five minutes and an initial accommodation period was observed. A Standardized Bolus Medium (SBM) kit (Trisco Foods Pty Ltd, Brisbane, Australia) which is made in accordance with the International Dysphagia Diet Standardization Initiative (IDDSI) framework (http://iddsi.org/framework/), was used to ensure standardized bolus viscosity and conductivity across different consistencies. SBMkit consists of apple flavoured sodium-chloride concentrate solution, a separate gum-based thickener (Precise Thick’N Instant), and an instruction sheet to make the appropriate IDDSI viscosity levels. A standardized HRIM protocol [15] was followed, with each participant performing triplicate cued swallow trials of nine bolus conditions (total 27 swallows) (thin liquid IDDSI 0, mildly thick liquid IDDSI 2, extremely thick liquid IDDI 4 with three volumes: 5 ml, 10 ml, 20 ml for each consistency). Each bolus was measured and administered to participants via a 20 ml syringe. Before any bolus administration, participants were informed to attempt single swallows per bolus where possible. There was a minimum 20 s break between swallows. On average, an HRIM session lasted 30–40 min.

All HRIM studies were performed by experienced clinicians who had certified training in HRIM, at the swallow research laboratory. The clinician labelled swallows during the procedure for later analysis. Any swallow variations (cough), adverse events, and the protocol completion rate were recorded. In compliance with ethical requirements, each participant was observed for 20 min after completion of the protocol. Once the protocol was completed, raw data was acquired at 20Hz (Solar GI acquisition system, MMS, The Netherlands). All the VFSS data collected based on a standard protocol [11] were rated for penetration-aspiration scale (PAS) [17] score by two expert clinicians.

Swallow gateway analysis

Pressure and impedance data were exported (ASCII format) and up-loaded (de‐identified) to the online Swallow Gateway™ platform (Flinders Partners Pty Ltd, Australia) for semiautomated analysis. Each participant’s pharyngeal swallows were analysed by a speech-language therapist who had completed the SwallowGateway analysis course. Uploaded data were analyzed using varied landmarks and validated metrics as described in Table 1 [4, 8, 16]. Analytic methods and reliability of SwallowGateway analysis have previously been described [4]. Some participants’ swallow data with recording errors (new equipment technical errors, swallow labelling error) were excluded from the final analysis. 50% of randomly selected HRIM data were analyzed by a second rater. Both raters were blinded to clinical characteristics. Markers placements were thoroughly reviewed by both the raters and difficult swallow analysis were resolved through consensus and referring to the expert tips.

Table 1 HRIM pressure flow parameters and definitions for core outcome metrics
Full size table

Statistical analysis

The average results for the triplicate swallows for nine bolus conditions (3 volumes and 3 consistencies) were tabulated in an Excel spreadsheet for statistical analysis using SPSS (IBM Corp., IBM Statistical Package for the Social Sciences (SPSS), v 27.0 Armonk, NY, IBM Corp). Individual values for each parameter was compared to Swallowgateway normative study data (www.swallowgateway.com) [14] and counted for abnormality if it was below 5th and above 95th percentile. As our study didn’t have a control group to compare the findings we selected swallowgateway norms as the values were derived from similar methods and equipment configuration to our study (same catheter, similar protocol and same bolus medium used for the oral trials). Association between disease severity and bolus characteristics was assessed using Spearman rank order correlation. Volume and viscosity main effects were evaluated using general linear mixed model repeated measures analysis. Bonferroni adjustment was applied to pairwise comparisons. A p-value of < 0.05 was considered to indicate statistical significance.

Results

All HRIM procedures were completed without complications. Twenty-one participants (5 females) completed HRIM with a mean age of 69 years (SD 7 years). The patient characteristics are shown in Table 2. Participants all demonstrated mild PD severity by the Hoehn and Yahr rating and PDQ-8 scores. The manometry procedure was well-tolerated by all the participants and no adverse effects were recorded across the cohort.

Table 2 Participant demographics characteristics and self-reported disease severity scores
Full size table

Dysphagia severity

All participants reported eating a normal diet with no modifications to food and drinks (IDDSI Level 7). However, 73% of participants (n = 15/21) reported swallowing disturbances and scored outside the normal range (> 3 points) on the EAT-10. In the VFSS examination, four participants aspirated (PAS 7–8) during the 100 ml straw drinking task (and two of them aspirated for 3 ml (P006), 20 ml (P036) and 3 cm3 (P036) paste bolus trials). The remaining seventeen participants had normal PAS scores for all swallow bolus trials. Self-reported PDQ-8 score was positively associated with self-reported dysphagia severity assessed by EAT-10 scores (rs (19) = 0.504, p = 0.02) (Fig. 1).

Fig. 1
figure 1

Positive correlation between Parkinson’s severity (PDQ-8) and dysphagia severity (EAT-10). Individuals score with a fitted regression line with 95% confidence interval

Full size image

HRIM parameters

IDDSI 0 bolus trials: compared to (below 5th and above 95th percentile) normative measures, participants demonstrated abnormally elevated mesopharyngeal contractile integral (n = 5/21), elevated hypopharyngeal contractile integral (n = 6/21), high UES integrated relaxation pressure (UES IRP) (n = 5/21) and low UES maximum admittance (UES MaxAdm) for 20 ml swallow trials. In other volume trials (5 ml, 10 ml) UES IRP (n = 7/21) and UES contractile integral (n = 5/21) were elevated in roughly a quarter to one third of participants.

IDDSI 2 and IDDSI 4 bolus trials: a greater number of participants had abnormally low UES MaxAdm with increased viscosity bolus (IDDSI 2: n = 10/21 and IDDSI 4: 8/21). Hypercontractility in the mesopharynx, hypopharynx, and elevated UES IRP was infrequent (n = 2, n = 3 and n = 4 respectively) for other consistencies. (All abnormal patients scores across volumes and viscosity are presented in Supplementary Table 1, and HRIM metrics means with 95% confidence intervals are presented in Supplementary Table 2.

Bolus volume and consistency

Both volume and viscosity increased UES maximum admittance (respectively F = 67.59, p < 0.001; F = 21.39, p < 0.001) (Table 3). The timing measure, distension to contraction latency (DCL), increased as the bolus volume increased (F = 31.93, p < 0.001). In contrast, bolus viscosity significantly reduced DCL (F = 11.50, p < 0.001). UES contractile integral demonstrated a significant correlation with bolus consistency, whereby reduced UES contractility was seen with increased viscosity (F = 6.80, p = 0.03). Bolus viscosity reduced the UES peak pressure (F = 5.25, p = 0.011). No other HRIM measures differed significantly in relation to increased volume and viscosity. Comparison plots are displayed in Fig. 2 for variables UES MaxAdm, DCL, UES PeakP, and UES CI.

Table 3 Summary of the main effects of volume and viscosity
Full size table
Fig. 2
figure 2

Graphs showing effects of volume and viscosity for a UES maximum admittance, b Distension to contraction latency, c UES Contractile integral, and d UES peak pressure

Full size image

No relationship between PDQ-8 severity and any of the HRIM measures was found. EAT-10 score was correlated with the following measures: UES basal pressure was significantly higher for (p = 0.04) thin liquid bolus trials in the group with elevated EAT-10 scores (≥ 7), velopharyngeal contractile integral was significantly lower (p = 0.04) for extremely thick liquid trials (IDDSI 4) in individuals with elevated EAT-10 scores and UES MaxAdm was significantly lower (p < 0.05) across most bolus conditions in the group with elevated EAT-10 scores (Supplementary Table 3).

Discussion

The present study explored pharyngoesophageal swallowing function in individuals with PD using HRIM metrics and characterized pharyngeal modulation of swallow parameters in relation to a wide range of bolus conditions. Key findings of the study were that individuals with PD presented with increased pharyngeal contractility (MCI, HPCI), abnormality in UES relaxation and distension (high UES IRP, low UES MaxAdm), and increased post-swallow UES contractility (UES CI). Both volume and viscosity changes had significant modulatory effects on UES maximum admittance and distension to contraction latency metrics, and UES CI and UES peak pressure measures were altered in relation to bolus viscosity.

In this current cohort of individuals with PD, self-reported dysphagia symptoms were prevalent in 73% of individuals but participants maintained unrestricted diets. Only four (19%) demonstrated abnormal penetration-aspiration scores during VFSS examination. No participant was being treated by a speech-language therapist for swallow dysfunction at the point of recruitment. This suggests that early swallow changes occur and are identifiable prior to individuals reporting or seeking dysphagia-related professional support. There was a significant positive correlation between self-reported Parkinson’s disease severity (PDQ-8) and self-reported dysphagia severity (Fig. 1). This finding is supported by other studies reporting advancing swallow impairment as PD stage advances [12, 19, 20].

Increased pharyngeal contractility in the mesopharyngeal and hypopharyngeal regions were exhibited in this cohort of PD patients, a finding mirrored by Szczesniak et al. in their recent study of 64 PD patients [21]. This may indicate an adaptive behaviour by the pharynx to achieve effective bolus transfer and prevent pharyngeal residue accumulation, given that we also found reduction in UES MaxAdm which represents restriction to flow across the UES. A recent study in normal adults, using a similar HRIM protocol to our own, by Ferris et al. [8] also demonstrated velopharyngeal, hypopharyngeal, and pharyngeal contractility increased during larger bolus volume trials and with increased bolus viscosity, suggesting that more intense force is generated in the pharyngeal chamber for propulsion of larger bolus volumes. Reduced force generation in the face of larger bolus volumes or increased bolus viscosity, due to weak pharyngeal muscles or muscle atrophy in the PD population, potentially results in inefficient pharyngeal constriction and less bolus impetus, slower bolus velocity and risks residue being left behind post-swallow. Several previous videofluoroscopic studies report reduced pharyngeal constriction as one of the key features associated with compromised swallow efficiency (residue) in patients with PD [1, 6, 19].

Ferris also reported adaptation of hypopharyngeal intrabolus distension pressure (IBP) for varied bolus conditions in neurologically healthy adults [8] but we did not identify this in our PD group. This may imply pharyngeal weakness in our cohort or loss of modulation skill or ability, given the lack of increase in pharyngeal contractility detected with presentation of increased volume and viscosity.

Elevated UES integrated relaxation pressure (UES IRP) was observed in our cohort, in keeping with previous study findings [21]. Resistance to bolus flow across the UES is influenced by the adequacy of UES relaxation, extent of UES opening (contingent on hyolaryngeal motion), and duration of UES relaxation [4]. Elevated UES IRP indicates abnormal flow resistance in these PD patients. Healthy adults demonstrated elevated UES IRP in response to increased volume and viscosity [8]. In our patients, although we saw an overall elevated UES IRP, this did not vary with differing bolus volumes or viscosities, but remained elevated throughout (Supplementary Table 1). Explanation for this may include that UES resistance may be a fixed aspect of the muscle or hyolaryngeal system, or represent loss of an adaptive response. If the UES had become fibrotic and non-compliant we would have expected solid bolus to also be restricted in passage. This consistency was not tested during this protocol, and we are unable to ascertain whether solid food transit impairment was present (although few participants complained of this). This may be relevant when considering intervention, as exercise therapy in this setting may need to be specific to hyolaryngeal movement, or may be less effective if a non-compliant muscle is present, in which case direct UES dilatation might offer greater benefit.

Interestingly, we observed that larger bolus volumes and increased bolus viscosity were associated with greater UES maximum admittance (UES MaxAdm) in this cohort, but that overall our participants demonstrated low UESMaxAdm. UES MaxAdm is a marker of the extent of UES opening, and low admittance values indicate reduced excursion of the UES producing increased bolus flow resistance [4]. The UES can also be forced open by hydrostatic pressure force if a large bolus is taken, if all other exits from the pharynx are closed (i.e. nasopharynx and airway). In this case, tongue pumping behaviour can drive a large bolus through the UES in a piecemeal fashion. In our study, the increase in UES MaxAdm (although not in the normal range) suggests the UES is still compliant but has not been distracted adequately (due to poor hyolaryngeal function) and the greater bolus volume is providing some hydrostatic force to drive bolus across the PES. There may still be an aspect of reduced compliance of the UES itself as connective tissue in the muscle becomes more rigid with age, and combined with lack of distraction related to reduced hyolaryngeal elevation, results in reduction in overall UES opening extent and duration. In PD, reduced oral-phase pressure generation due to bradykinetic lingual movement, premature spillage due to reduced tongue posterior bolus containment, delayed swallow reflex, increased pharyngeal transit time, and reduced hyolaryngeal excursion may be contributing to poor bolus impetus and reduced UES opening. These findings have been reported in videofluoroscopic swallow studies in PD previously [1, 17]. A combination of factors is likely to emerge in each individual, that dictates overall pharyngeal transit. Effects of bolus volume and viscosity on UES MaxAdm evident in this cohort are consistent with previous work [22] and also with values in neurologically healthy individuals where UES MaxAdm varied across volumes and viscosity [8].

Post-swallow UES contractility was abnormally elevated in a greater number of PD patients for 5ml thin liquid bolus trials, compared to larger volumes and thicker bolus trials. Potential explanations for these differences include an immediate post-swallow contraction to help prevent retrograde flow of thin bolus, which might otherwise easily regurgitate, if muscle weakness is present, or that if the pharynx receives a liquid volume bolus, it requires greater concentric clearing contraction to reach it (as liquid hugs the pharyngeal contours) and provide propulsion compared to a larger, less free-flowing and more cohesive bolus, where pharyngeal wall constriction is able to contact the bolus at an earlier timepoint. These mechanisms may also explain elevated UES peak pressure for thin bolus compared to thick. The lack of change when different volumes were presented is interesting, and may reflect gravity influence on thin liquids wherein the natural velocity of the fluid is greater or just the capacity of the pharynx overall to manage volumes within the 5–25 ml range. The hydrostatic pressure generated by a larger liquid bolus size would also increase liquid velocity through inertia and thus the pharynx may not need to apply additional pressure to the larger liquid bolus. Interestingly, neurologically healthy adults did not alter UES contractility or UES peak pressure in relation to bolus viscosity possibly suggesting redundancy in the ability of the normal pharynx to generate pressures across a range that enables consistent flow rates through the UES [8, 13].

Our study shows that pharyngeal distension to contraction latency (DCL) significantly altered in relation to both bolus volume and viscosity. A longer duration of DCL was seen with increased volumes and a shorter duration of DCL was observed with increased viscosity, in line with previous study findings in normal adults [8, 13]. Greater volume bolus will take longer to distend the pharynx as it fills, increasing the DCL whereas a more viscous but same-sized bolus will require more impetus to travel the pharynx and through the UES in the same timeframe, therefore contraction is occurring earlier to ensure flow rate consistency and bolus transit times remain normal. Taken together, these findings suggest that pharyngeal swallow sensory-motor regulation is still preserved in the early stages of PD.

Study limitations and future directions

This study described preliminary findings of manometrically characterized pharyngoesophageal swallow changes in a small cohort (n = 21) of early/mid-stage PD patients. Although we aimed to recruit participants with a range of PD severity stages, study data collection was affected by the COVID-19 pandemic. Therefore, our findings cannot be generalized to more severe stages of PD. Future studies should include a larger sample size with a wider range of disease severity. This study did not exclude patients with normal EAT-10 scores. However, future larger group studies should consider including more homogeneous groups either with or without complaint of swallowing difficulties in order to provide stronger analysis. We presented bolus conditions with varied volumes and viscosity in a fixed order rather than providing it in random order. Hence, data may have been influenced by time-order and fatigue effects. We haven’t included an age-matched control group, however, we were able to compared our HRIM measures with normative data provided by the open-source SwallowGateway web-based application which utilizes the same hardware as our laboratory. This application includes data across all ages (≥ 18 years) and may not be specifically generalizable to older people.

Conclusions

Using HRIM evaluation we were able to describe changes in pharyngeal swallow behaviour and responses triggered by altering bolus volume and viscosity conditions in PD patients. Key swallow modulatory changes in PD patients include greater resistance to flow (decreased UES MaxAdm and elevated UES IRP) with corresponding increased pharyngeal contractility and elevated IBP. Lack of responsiveness of UES IRP and IBP to bolus parameter change, suggests either, that hyolaryngeal movement was impaired and unable to provide normal UES distraction or that fixed resistance may be present at the UES contributing to increased resistance. Implications of these findings when planning management suggest that exercise-based therapy alone may not address early swallow changes and that targeted therapy at the UES may also have a role to play in reducing resistance to bolus flow. Use of HRIM evaluation in PD individuals would help guide the choice of treatments, and may provide further insights into PD-related dysphagia to help optimize the diagnostic and treatment planning framework.