Introduction
Oropharyngeal dysphagia (OD) is a common non-motor symptom in idiopathic Parkinson’s disease (PD) [
1,
2]. The burden of OD in PD is immense, as it affects health-related quality of life (QoL) [
3‐
5], and may lead to complications such as aspiration pneumonia [
6,
7]. Even in an early PD stage, patients report a lower swallow-related QoL compared to healthy control subjects [
3]. However, patients with a moderately advanced PD disease stage do not report the worst swallow-related QoL compared to the patients in the early disease stages.[
3]. It appears that the swallow-related QoL only decreases further when the patients are in advanced Hoehn and Yahr (H&Y) stage [
3]. This suggests that the decline of self-report swallow-related QoL stagnates despite the progression of PD during its early H&Y stages or that PD patients develop compensatory swallowing strategies or coping mechanisms that inhibit the decline of swallow-related QoL despite the progression of OD. In case of the latter, an inconsistency is expected between the patient reported outcome measures (PROMs) and the investigator reported outcome measures (IROMs) on swallowing.
The inconsistency between patient self-report swallow-related QoL and the actual swallowing function using fiberoptic/flexible endoscopic evaluation of swallowing (FEES) or videofluoroscopy of swallowing (VFS) has been described in several studies on OD due to other underlying disorders such as acute stroke, myotonic dystrophy and head and neck cancer [
8‐
11]. Silent aspiration occurs in about 20% of the PD patients and is one of the main risk factors for developing aspiration pneumonia [
12,
13]. For PD, the relationship between the results of validated self-report swallow-related QoL questionnaires and of instrumental tools such as FEES or VFS has not been reported in the literature before.
Therefore, the objective of the present study was to determine the relationship between PROMs and IROMs on swallowing in PD patients. To further explore the characteristics of this relationship, clinically relevant subgroups of patients within the study population, based on similar PROMs and/or IROMs, were identified and studied.
Discussion
The objective of the present study was to determine the relationship between PROMs and IROMs on OD in PD. Only a relationship with a moderate agreement (AUC = 0.6–0.7) between the PROMs and IROMs on OD in PD was found. This suggests that there is some sort of inconsistency between the signs of OD identified by clinicians using FEES and/or VFS and patient self-report swallow-related QoL questionnaires.
This inconsistency between PROMs and IROMs is not new in the literature on neurogenic dysphagia. In a cohort of 119 PD patients, Nienstedt et al. found that only 50% of the patients with severe aspiration (Penetration Aspiration Scale score > 6 [
30]) during FEES reported swallowing complaints in the relevant domains of the Unified Parkinson Disease Rating Scale (UPDRS) II and in the non-motor symptoms questionnaire (NMS). The majority of these patients described their difficulties as ‘slight restrictions in swallowing’ [
32]. Pflug et al. used a single question to evaluate whether PD patients experienced swallowing impairment and compared this outcome to signs of OD using FEES [
33] Only 6% (
n = 5/119) of the PD patients showed a normal pharyngeal swallow during FEES. However, 73% (
n = 87/119) denied any swallowing impairment. The majority of the PD patients without OD complaints showed pharyngeal pooling of dyed water (52%;
n = 45/87), bread (93%;
n = 81/87), and biscuit (86%;
n = 75/87) and 16% (
n = 14/87) showed aspiration [
33]. Only 12–27% of the PD patients with signs of swallowing impairment during FEES reported swallowing complaints [
32,
33]. The current study also showed a moderate agreement between PROMs and IROMS in dysphagic PD patients. However, it is important to emphasize that previous studies described PROMS using OD symptom and FHS questionnaires and did not report on swallow-related QoL questionnaires.
To further elaborate this moderate agreement between PROMs and IROMs in the present study, a two-step cluster analysis was performed. It was hypothesized that there are clusters of patients with similar outcomes on FEES or VFS resulting in similar clinical patient labels, but with different outcomes on MDADI or DSS scores. The cluster analysis could help to understand why some PD patients with similar signs of OD during FEES or VFS have swallowing complaints and others don’t. Using the two-step cluster analysis, patients of cluster 1 in the
glossopalatal label (Fig.
1) showed signs of OD during FEES and/or VFS and at the same time the lowest mean MDADI subscale and DSS scores, representing a poor swallow-specific QoL. However, the clinical patient label
glossopalatal also contained patients of cluster 2 who had the highest mean MDADI subscale and DSS scores (highest swallow-specific QoL), and signs of OD on the IROMs. In the attempt to identify confounders that could predict the differences in the level of swallow-specific QoL presented by patients in cluster 1 and 2, patient characteristics were added to the analysis. However, the variables age, gender, H&Y scale, and the score on the other clinical patient labels could not be identified as confounders. The exact reason for the significantly different mean scores on the PROMs in patients with similar IROMs was therefore not found in this study.
A similar result was seen for the clinical patient label pooling. Only two clusters were found: one with signs of postswallow vallecular and/or postswallow pyriform sinus pooling and the other without. Interestingly, the mean MDADI subscale and DSS scores did not significantly differ between both clusters. Apparently, the level of swallow-specific QoL did not seem to depend on the presence or absence of pharyngeal pooling.
There are numerous hypotheses regarding the pathophysiology of OD in PD. Different sites in the nervous system may be affected [
34]. A possible explanation for the inconsistency between PROMs and IROMs on OD in PD may be that the different sites of pathology in the nervous system may affect the swallowing function and the subjective perception of this in a different way. So, the phenotype of OD of an individual PD patient seems to encompass more than just the biomechanical swallowing function measured by IROMs. The OD phenotype includes the dimension of ‘the subjective perception of the swallowing disorder by the patient’ as well. PROMs and IROMs really seem to represent different dimensions of OD that together determine an OD phenotype in an integrated manner. The most well-known hypothesis of the pathophysiology of OD in PD is the lack of dopamine in the basal ganglia [
35]. Functional magnetic resonance imaging (fMRI) studies in healthy subjects showed increased activation in parts of the basal ganglia namely the globus pallidus and putamen during swallowing [
36]. Restoring the dopamine levels in these areas using dopaminergic medication or deep brain stimulation seemed to significantly improve swallowing in some PD patients [
37]. However, several studies showed no significant improvements or worsening of OD using dopaminergic medication or deep brain stimulation, suggesting that there are different pathophysiological mechanisms in developing OD [
34,
37]. Another site of pathology in PD are the non-dopaminergic pathways which might be affected by the development of Lewy bodies. Lewy bodies are abnormal aggregations of mainly alpha-synuclein proteins and are related to neuronal cell loss [
38]. These Lewy bodies appear in the brainstem and cortex as PD progresses and were found in important pathways related to swallowing such as the dorsal nuclei of the glossopharyngeal and vagal nerve [
39]. Lewy bodies were not only found in the central nervous system, but also in the enteric nervous system, and in the sensory and motor nerves of the pharyngeal wall [
40,
41]. A possible hypothesis is that these different sites of pathology relate to different phenotypes of OD in PD, and require different diagnostic and therapeutic approaches.
Besides the different sites of pathology which may relate to different phenotypes of OD in PD, the occurrence of compensatory mechanisms may be another attribute to the different phenotypes. Some PD patients develop compensatory mechanisms that prevent them from having swallowing complaints [
42]. Using magneto-encephalography (MEG) a shift in cortical activation during swallowing was found from the affected supplementary motor area to the lateral motor, premotor, and inferolateral parietal cortices in PD patients without clinical signs of OD. PD patients with clinical signs of OD did not show this shift on MEG [
42]. Next to this compensatory shift in cortical activation several other compensatory strategies such as bolus modification and volume adjustment by taking smaller sips or bites may spontaneously be developed by PD patients [
43]. This may improve patient’s self-perception of swallowing, and also the safety and efficiency of swallowing, but does not necessarily improve the biomechanics of their actual swallowing disorder.
Multiple reasons may underlie this moderate agreement between PROMs and IROMs on OD in PD. The absence of a support network, the level of cognitive impairment, or the presence of neurobehavioral conditions such as mood disorders or optimism may affect a patient’s perception of swallowing [
44]. This study highlights that there are PD patients with similar FEES and/or VFS findings that cannot be lumped together under the same pathophysiological umbrella due to their differences in PROMs. This research has an important clinical relevance since it can give rise to differentiations in OD management for PD in the future.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.