Oropharyngeal dysphagia (OD) is recognized as a critical geriatric syndrome that significantly diminishes the quality of life in older patients and is associated with several adverse outcomes, including hospitalizations, pneumonia, malnutrition, dehydration and increased mortality (Baijens et al., 2016; Banda et al., 2022).

The etiology of OD is complex and depends primarily on the presence of neurological diseases and age-related changes in multiple swallowing domains including sensory function and disease related changes of anatomical structures (Rajati et al., 2022; Abu-Ghanem et al., 2020). One of the neurological diseases that is frequently associated with dysphagia is dementia. Contrary to extensive research that has been conducted on dysphagia in stroke patients, where dysphagia may diminish over days to weeks, less is known about the dynamic in neurodegenerative diseases that typically show a progressive decline (Parlak et al., 2022; Sheikhany et al., 2019). As the condition advances in patients with dementia, there is often a notable decrease in pharyngeal clearance and opening of the upper esophageal sphincter, accompanied by frequent occurrences of penetration and/or aspiration (De Stefano et al., 2020; Mira et al., 2022). These changes exacerbate the risk of severe health consequences, underscoring the need for precise diagnostic and management strategies.

Flexible endoscopic examination of swallowing (FEES) is a well-established, reliable, bedside diagnostic tool for assessing swallowing disorders in geriatric patients (Dziewas et al., 2024). Recent advancements include FEES-based classification of dysphagia phenotypes (Warnecke et al., 2021). In recent studies, the use of FEES mainly focussed on geriatric patients with cognitive impairment resulting from cerebrovascular accidents and pneumonia (Benjapornlert et al., 2020), presbyphagia (Labeit et al., 2022) or rare neurodegenerative diseases such as amyotrophic lateral sclerosis (Tye et al., 2021). In particular, characteristic dysphagia patterns have been associated with Parkinson’s disease, stroke and neuromuscular diseases (Warnecke et al., 2021). Nevertheless, the specific swallowing pathologies associated with Alzheimer’s disease and other forms of dementia remain to be established, and the impact of dementia on dysphagia is still poorly understood. Accordingly, the objective of the present study was to provide a systematic analysis of the endoscopic pathologies associated with different types of dementia and concomitant dysphagia.

Baseline characteristics of study participants are presented in Table 1. Of the 313 geriatric inpatients with dementia who were screened between 2019 and 2024, 107 patients with endoscopically confirmed oropharyngeal dysphagia were included in the final analysis. Reasons for exclusion included missing or incomplete FEES (n = 87), predefined exclusion criteria such as delirium or infection (n = 60), and an absence of pathological FEES findings (n = 59). A detailed flowchart of patient selection is provided in Supplementary Figure S1. The age of the study population ranged between 65 and 96 years (82.5 ± 6.7 years), with 54% being females. Major reasons for hospital admission included falls and fractures, immobility after surgery cardiovascular diseases, malnutrition, heart failure and neurodegenerative diseases. The Barthel index recorded a median score of 40, indicating a moderate to severe dependency in activities of daily living among the participants. Furthermore, 80 patients (74.8%) were identified as having probable sarcopenia according to the Sarc-F score and frailty was observed in 97 patients (91.5%) based on Clinical Frailty Scale.

Table 2 summarizes the prevalence and severity of different dementia types and dysphagia pathologies as assessed by FEES. The cohort comprised 107 patients, with one-third of each dementia type including Alzheimer’s disease, mixed-type dementia, i.e., Alzheimer’s disease with vascular lesions and Parkinson’s-related, i.e., Parkinson’s disease with dementia and or possible diagnosis of Lewy body dementia (McKeith et al., 2017). Notable dysphagia pathologies included a high prevalence of residue in valleculae (96%), impaired white out (75%), and a complex dysphagia pathology (72%), with aspiration being the least common condition observed in 13%.

Table 3 provides a comparative analysis of dysphagia pathologies across three dementia types. Significant variation was noted in the prevalence of penetration in the complex dysphagia pathology. Penetration was significantly more frequent in Parkinson’s-related dementia patients (63%; p = 0.016), compared to Alzheimer’s disease (31%) and mixed-type dementia (38%). Complex dysphagia was also most prevalent in Parkinson’s-related dementia (90%; p < 0.010), and distinctly higher than in Alzheimer’s dementia (59%) and mixed-type dementia (65%). The most common pathology was residue in the valleculae with an equal distribution between dementia types. In contrast, aspiration was least prevalent, with a non-significant variation across types, indicating limited clinical utility for differentiating dementia types. These findings are illustrated by Figures 1, 2. Figure 1 shows the prevalence of the distinct endoscopic findings in the three dementia groups and Figure 2 the same from the perspective of the dementia type.

Partial correlation analyses indicated a significant association between impaired white out and aspiration among various dysphagia pathologies (r = 0.283, p < 0.001). No other significant correlations were identified between different dysphagia pathologies. In addition, we conducted a regression analysis to investigate the independent effects of various factors, including age, gender, severity of dementia, sarcopenia, handgrip strength, Barthel Index, and frailty, on dysphagia severity. The analysis revealed that none of these risk factors demonstrated a significant association with the severity of dysphagia as evidenced by the following p-values: severity of dementia (p = 0.926), sarcopenia (p = 0.836), age (p = 0.267), gender (p = 0.304), handgrip strength (p = 0.174), Barthel Index (p = 0.609), and frailty (p = 0.737).

The etiology of dysphagia is multifactorial, involving both cognitive impairments associated with dementia and behavioral disorders. These behavioral disorders may include buccofacial apraxia and refusal to eat, resulting in a prolonged oral phase and frontal bolus spillage. Additionally, inter- and post-deglutitive pathomechanistic changes can be observed using FEES. Only few studies have focused exclusively on employing FEES to explore dysphagia in patients with Alzheimer’s disease compared to various stages and forms of dementia (Parlak et al., 2022; Sheikhany et al., 2019; De Stefano et al., 2020; Shapira-Galitz et al., 2019; Suh et al., 2009). This study extends the current understanding by providing insights into the prevalence and characteristics of various dysphagia pathologies across different types of dementia using FEES. Bolus residuals in the vallecula and piriform sinus was detectable in almost all dementia patients (96%). This high prevalence exceeds rates reported in prospective age cohorts, underscoring that dementia-specific neurodegenerative changes significantly contribute to residual function in addition to age-related sensory loss (Labeit et al., 2022).

Our findings highlight significant variation in swallowing dysfunction and findings, with a notably higher prevalence of bolus penetration and complex dysphagia pathology observed in patients with Parkinson’s-related dementia. Our results are consistent with those of a previous study in a relatively large cohort of Parkinson’s disease patients studied with FEES. Similar to our cohort, bolus residues were observed in 93% and bolus penetration was observed in 55% (Pflug et al., 2018). Theories suggest that food residue accumulation, depending of their presence in the valleculae or the sinus piriformis, and bolus penetration could be due to reduced tongue base retraction, velopharyngeal closure, weaker pharyngeal muscles, and a decrease in upper esophageal sphincter opening (Taira et al., 2021; Higo et al., 2001).

Another notable finding of our study is that the most common dysphagia pathology observed across all types of dementia was the presence of residues in the vallecula. In line with our results, bolus residue in the vallecula and the sinus piriformis were also the most prevalent findings in another FEES-based study on Alzheimer’s disease patients (Parlak et al., 2022). Previous research has linked vallecular residues in dementia patients to weakened pharyngeal muscle strength and a delayed swallowing reflex, both of which are associated with an increased risk of aspiration (Shapira-Galitz et al., 2019; Molfenter and Steele, 2013). Contrary to expectations, our results did not indicate a significant association between the presence of residue and an increased rate of aspiration. However, we did observe a correlation between reduced pharyngeal muscle strength, quantified by the white-out phenomenon, and a higher frequency of aspiration in the comparative analysis across groups. The white-out phenomenon has been proposed as a semi-quantitative indicator of pharyngeal contractility (Labeit et al., 2021a) and pharyngeal contractility has been identified as a reliable predictor of aspiration in both FEES and videofluoroscopic swallowing studies (Miles and Hunting, 2023; Ku et al., 2021). This highlights the importance of measuring pharyngeal muscle dynamics as a component of comprehensive dysphagia assessment in dementia patients, providing critical insights into the mechanisms underlying swallowing disorders in this population.

The findings of our study indicate that there are no differences in dysphagia pathologies between patients with Alzheimer’s dementia and those with Alzheimer’s dementia and concomitant small vessel disease. Comparative analyses of these two forms of dementia are lacking in the literature, although an older study employed videofluoroscopy to examine the differences in dysphagia between patients with Alzheimer’s disease and those with vascular dementia (Suh et al., 2009). With regard to the characteristics of dysphagia, no relevant differences were observed in the prevalence of bolus residue in Alzheimer’s patients and patients with vascular dementia (80.0% vs 67.6%, p = 0.502), neither in the occurrence of penetration (73.3% vs. 55.8%, p = 0.248), which is in line with the data in our study. However, in comparison to our cohort, there was a prolongation of the oral phase in patients with Alzheimer’s disease (p = 0.008) and a higher incidence of aspiration in patients with vascular dementia (p = 0.024). It is important to note that the Alzheimer’s patients in the above-cited study showed significantly greater cognitive impairment than those in our cohort. Previous literature has indicated that as dementia progresses, the prevalence of behavioral eating disorders, such as buccofacial apraxia, tends to increase (Seçil et al., 2016). Due to this inter-individual variability of neuropsychiatric symptoms, the reported prevalence of dysphagia in Alzheimer’s disease spans from 2.4% to 100% (Mira et al., 2022).

In summary, no specific endoscopic pathologies of dysphagia could be detected in patients with dementia, except for minor specifics in the Parkinson’s group. The reasons for this may be found in the overlap of the cerebral swallowing network with the neuropathological spread of beta-amyloid pathology in Alzheimer’s disease and the pattern of cerebral microangiopathy. Cortical regions associated both with the swallowing network and the trajectory of neurodegeneration include the lower frontal gyrus, the anterior cingulate cortex, the orbitofrontal cortex, and the supramarginal gyrus (Cheng et al., 2022; Braak and Braak, 1991; Thal et al., 2006). A review of cerebral microangiopathy in patients with dysphagia revealed that white-matter lesions, particularly in the internal capsule, corona radiata, insula, and corpus callosum, were linked to swallowing disorders (Alvar et al., 2021). The lack of differences in the swallowing pathologies of patients in the FEES may be attributed to the fact that Alzheimer’s disease and cerebral microangiopathy progress along a continuum and overlap with areas of the cerebral swallowing network.

For clinical practice, our study shows that patients with dementia have impaired bolus clearance. The relatively rare occurrence of white-out cannot be identified as a reliable indicator of this due to the lack of standardization in recording, evaluation, and interpretation. Individualized dysphagia therapy should therefore continue to be decided on an individual basis for patients with dementia, for example, through meal assistance or careful adjustment of food consistency (Lueg et al., 2024). Our study has several limitations. Conducted as a retrospective analysis with a limited number of participants, generalizability of data is possibly constrained. In addition, the multimorbidity of the patients does not allow for a clear determination of causality between the pathophysiology of dementia and specific types of swallowing disorders. Furthermore, the clinical relevance of dysphagia, particularly in patients with Alzheimer’s disease, is likely influenced by kinematic aspects of swallowing alongside cognitive and behavioral symptoms - factors not fully captured due to the retrospective nature of our study. In future studies, a structured behavioral swallowing assessment, including validated screening for buccofacial apraxia and standardized dysphagia questionnaires administered by speech therapists, would be necessary to differentiate behavioral components of swallowing impairment in dementia. One strength of the study is the utilization of FEES in patients with dementia which is a challenge due to reduced cooperation of the participants. To gain further insight into the influence of dementia on swallowing disorders, it would be beneficial to conduct a comparative analysis with our collective and geriatric patients who have been diagnosed with mild cognitive impairment and Parkinson’s disease without concomitant dementia. Such studies will be important for developing and implementing effective therapies tailored to disease stages and individual dysphagia patterns, focused on enhancing pharyngeal muscle strength and optimizing nutritional interventions.

Despite small differences in participants with Parkinson’s-related dementia, our study did not identify a typical dysphagia pathology in different types of dementia. Therefore, the individual presentation of dysphagia must be determined in each patient to enable a tailored intervention.