Introduction

It is known that an important group of patients with oropharyngeal dyspaghia (OD) is reliant on texture modified food (TMF) [1, 2]. Moreover, provision of TMF is described as a cornerstone of dysphagia management [3]. Nevertheless, some remarks can be made concerning this widespread application. Firstly, the clinical decision concerning diet recommendation is ideally made by a multidisciplinary swallowing team based on a standardized observation during swallowing a limited number of boluses, generally of fluid, semisolid, and solid consistencies, or, using the International Dysphagia Diet Standardization Initiative IDDSI level 0 and IDDSI level 3 or level 4, and sometimes IDDSI level 7 [3, 4]. According to the detailed definitions described in the IDDSI-framework, no particles – described as bits - might be present in IDDSI level 3 or lumps in IDDSI level 4 [5]. From IDDSI level 5, the presence of particles with a particle size less than 4 mm width and 15 mm in length is allowed, if they are easy to squash with the tongue [5]. However, if persons lack skills for particle dimension by chewing or mashing with the tongue, they should be served a texture that is ready-to-swallow with minimal additional work. Persons with decreased swallowing safety and/or efficiency are frequently put on a diet of homogeneous pureed food, corresponding to IDDSI level 3 or level 4 without considering pureed food containing particles (i.e. IDDSI level 5).

Based on clinical findings and observations that particles affect chewing and bolus formation which subsequently decreased safety and efficiency of swallowing, OD-patients are often recommended to restrict their functional oral food intake to merely homogeneous, pureed foods (i.e. IDDSI level 3 or level 4) [1, 2]. However, the use of TMF is mainly based on theoretical concepts of improving safety and efficiency during swallowing, with low evidence for this adaptation [6, 7]. In addition, TMF is often prepared with very limited resources and a lack of knowledge of the kitchen department, resulting in meals considered as unattractive, always looking the same, and tasteless by the patients with OD [8, 9]. Consequently, adherence rates are low, increasing the risk of malnutrition and aspiration [3, 6, 9,10,11,12,13,14]. However, enhancing taste, appearance and presentation of TMF improves resident quality of life (QoL), dining experience and weight status [8,9,10,11,12,13,14].

The adjustment from a standard oral diet towards TMF is expected to change sensory bolus input driving the swallowing function. Steele & Miller [15] described in their review the influence of sensory information on all phases of swallowing. During the oral phase, the tongue and palate have specific receptors providing sensory input over multiple fibers about bolus texture, shape, and size in order to adapt lingual forces to transport the bolus efficiently through the pharynx [16]. During the pharyngeal phase, sensory input triggers the swallowing reflex and modulates the sequential motor activity of muscles transporting the bolus through the pharynx. During the esophageal phase, sensory input modifies secondary peristalsis [15]. Consequently, modifying bolus characteristics as volume, viscosity and temperature, have been shown to influence sensory input and therefore the entire swallowing function. For example, expanding bolus volume increases tongue strength during swallowing and results in shorter latencies to evoke the swallowing reflex [17,18,19,20,21]. Higher viscosity and consistency have been correlated with oropharyngeal transit time and intrabolus pressures resulting in higher values for tongue strength during swallowing, increased hyoid movement, and improved UES opening [17, 19, 20, 22,23,24]. Finally, changing temperature seems to be effective in triggering voluntary-induced swallowing and decreases oral transit time [25, 26].

Based on this knowledge concerning the influence of sensory input on swallowing, we could hypothesize that adding particles into TMF might influence sensory pathways as well. Nevertheless, research concerning the impact of bolus particles on tongue strength or swallowing is scarce. The majority of publications are focusing on the rheological properties and sensory evaluation of TMF, applied in healthy participants [27, 28]. Tobin et al. [27] performed research in 19 healthy elderly (aged 61–81) on the relationship between the perception of ease of swallowing and rheological parameters of particulated foods. The particle size distribution, indicating the percentage of particles of a certain size, varied between 221.3 μm and 409.7 μm, depending on the preparation process. They concluded that suspensions with a broader particle size distribution and smaller particle size required less perceived effort, with a good agreement between particle size measurements and particle size perception by the panel [27]. Thus, besides changing bolus volume, viscosity and temperature, manipulating bolus particles is expected to influence swallowing function as well. Moreover, adding particles in TMF indicates a more normal diet and a diet at a higher IDDSI-level. This adaptation might ameliorate attractiveness, taste and thus food related QoL and nutritional status. Nevertheless, as far as we know, research concerning the impact of adding particles on tongue strength swallowing pressures, swallowing safety and swallowing efficiency in patients with OD has never been analyzed before.

The aim of this proof-of-concept study is to demonstrate the feasibility of adding particles in TMF on the swallowing process of patients with OD, especially on tongue pressure, safety and efficiency.

Methods

Participants

Ten adult participants with OD were included by means of convenience sampling at the Antwerp University Hospital were they were all treated for dysphagia. Inclusion criteria were (1) dysphagia based on MASA-score below 178 [29] and (2) Dutch as maternal language to allow comparison with normative data for tongue strength [30, 31]. All included patients reported absence of premorbid taste disturbances such as dysgeusia, ageusia, hypogeusia, phantogeusia or parageusia. No patient was prescribed treatment or drugs with a known side effect on taste perception and food intake. The only exclusion criterium was (treatment for) head and neck cancer since a high prevalence of dysgeusia, ageusia and hypogeusia in this population is known to influence the swallowing function [32]. Presbyphagia is defined as a swallowing disorder without any specific etiology [33]. Participant characteristics including premorbid level of food are described in Table 1. Current food intake was described by their reported functional oral intake as measured by the Functional Oral Intake Scale [34].

Table 1 Participant details
Full size table

Study Design

For this cross-sectional prospective trial, all participants accomplished two assessments: (1) A clinical examination of tongue strength by an experienced speech language pathologist (SLP). (2) A clinical examination of swallowing function by means of a fiberoptic endoscopic evaluation of swallowing (FEES) by an experienced ENT-physician and experienced SLP in order to evaluate swallowing safety and efficiency. Both assessments were executed consecutively during the same day with waiting time of less than one hour and no medication or medical procedures in between.

Instrumentation

The Iowa Oral Performance Instrument version 2.2 (IOPI Medical LCC, Redmond, WA USA) was used for tongue strength measurements. The location for anterior tongue strength measurement was determined by placing the part of the bulb closest to the connection tube in contact with the posterior face of the upper incisors, thereby positioning the bulb against the hard palate just posterior to the alveolar ridge [34, 35].

FEES was performed using a Olympus CLV-180® nasofibroscope coupled to an Evis Exera II Xenon light source, model CLV-180. The audio and images of the entire examination were visualized on the monitor during the examination and recorded and stored by the software application Audiqueen (OtoConsult NV, Antwerp, Belgium) to allow for analyzing afterwards.

Bolus Characteristics

Concerning the test samples, 3 different types of boluses were designed and extensively tested. In lab conditions, puree and mechanically millimetered cube shaped particles (2mm3 and 4mm3) of fresh carrots were separately prepared and then merged into a ratio of 70%/30%. No thickening agents were used in the entire preparation process. A panel of three chefs gastro-engineering has, separately from each other, tested all preparations for taste properties as well as IDDSI level 3 and IDDSI level 4 properties. All testing was repeated three times and was always performed before and after the addition of blue dye, as well as the cooling and controlled warming of the different purees. No significant differences were determined in terms of taste or consistency. Subsequently, a series of each type of bolus was prepared, and individually packaged under strict hygienic conditions, for cool (non-frozen) storage and transport. Bolus 1 (B1) consisted of 7 ml IDDSI level 4 carrot puree (5 kPa compressive strength). Bolus 2 (B2) consisted of 7 ml B1 enriched with 30% small particles of carrot (2 × 2 × 2 mm, 10 kPa compressive strength). Bolus 3 (B3) consisted of 7 ml B1 enriched with 30% larger particles of carrot (4 × 4 × 4 mm, 10 kPa compressive strength). The bolus volume of 7 ml was preferred to increase similarity with natural eating habits, since natural sip volumes of similar consistencies are described ranging between 6.1 and 7.9ml [36]. Each bolus existed in two variations, carrot puree with a natural orange color version for the SLP-examination and a blue-died version to enhance visualization of pharyngeal bolus transit during the FEES-examination [37]. The boluses required for each examination were heated in a controlled manner so that they reached 40 °C upon application. It is described that the texture of these semi-solid foods allows to be squeezed between the tongue and palate, and thus requires a minimal amount of masticatory cycles [38]. Since the compressive value of the particles did not exceed the compressive value of the pureed part of the bolus, current boluses are hypothesized to require a minimum amount of mastication as well.

Outcome Measures

Firstly, tongue strength was assessed during the SLP examination by measuring (1) regular effort bolus swallows, named PswalB (2x PswalB1, 2x PswalB2, 2x PswalB3), (2) regular effort saliva swallows (3x PswalSAL), and (3) anterior maximal isometric pressure (3x MIPA). This specific order was preferred clinically to prevent a possible impact from the maximal effort task when measuring MIP on the regular effort tasks PswalB or PswalSAL. In order to measure PswalB, the participant placed a spoon in their mouth, filled with the natural-colored orange bolus. Subsequently, the researcher placed the IOPI-bulb on top of the bolus at the same location of the MIPA measurements [30]. The order of the three boluses was randomly determined per participant by an online randomization tool (randomizer.org). The participant was asked to ‘Swallow the bolus as you would do in normal circumstances.’. MIP values were registered by the IOPI for two subsequent swallows and the mean value of both was calculated as PswalB. PswalSAL was a regular effort saliva swallow as described by Steele [39] and the mean MIP-value of 3 subsequent swallows was retained for further analysis as PswalSAL. Finally, MIPA was determined as the best MIP value out of 3 as described by Lazarus et al. [40].

Secondly, swallowing safety and efficiency were judged by quantifying penetration, aspiration, and pharyngeal residue during swallowing the blue-died boluses B1, B2 and B3. Since these measurements fit in a proof of concept study, all boluses were presented in a fixed order based on the hypothetical risk on aspiration: B1 followed by respectively B2 and B3. In case of medical concerns from the multidisciplinary team during the procedure, the FEES-examination was discontinued. For each participant, all boluses were clipped separately resulting in 30 recordings. These 30 recordings were randomized by a researcher blinded to this study. Two standardized and validated scales were judged by 3 blinded raters (2 SLPs and 1 ENT), relying on expert consensus based on the recorded images during FEES-examination. The Penetration and Aspiration scale (PAS-scale [41]) is an 8-point scale used to characterize the depth and response to airway invasion [42]. Value 1 reflects a normal swallowing function, i.e. no entry of material into trachea or larynx, and value 8 describes the worst degree of aspiration, i.e. silent aspiration. The Pooling score [43] was used to measure residue and consists of 3 subscales ‘site’, ‘amount’ and ‘management’. The final score consists of the most severe condition for each parameter per bolus. The higher the score, the more severe the dysphagia, due to the presence of residue.

Data Analysis

Sample size was calculated with PAS as primary outcome variable by means of Gpower [44], based on the results of Troche et al. [45] pointing out significant differences in PAS when varying bolus consistency (p = .007). In conclusion for an effect size of 0.5512, alpha error 0.05 and power 0.8, a sample size of 9 participants was calculated. In order to accommodate for drop-outs, the targeted total sample size was 10.

Statistical analysis was performed by means of SPSSv21. Descriptive statistics, tests for normality (Kolmogorov–Smirnov) and RM Anova were performed for analyzing the impact of adding particles on tongue strength. Differences on PAS and Pooling score between the 3 boluses were calculated by means of a Friedman test. The statistical significance level was set at 0.05.

Ethical Approval

This study was independently reviewed and approved by the Committee for Medical Ethics UZA-UA (B3002021000008). All subjects agreed voluntarily to participate in this study and signed an informed consent in full accordance with ethical principles including the World Medical Association Declaration of Helsinki (version 2002) and additional requirements.

Results

Tongue Strength (TS)

There is no significant impact of bolus type on TS during swallowing B1, B2 or B3, as demonstrated in Fig. 1 (F(1.393, 12.536) = 0.524, P = .541-). Further analysis shows no significant differences between PswalSAL and PswalB (F(1.393, 12.536) = 0.524, P = .541-). Mean functional reserve [39] defined as the ratio Pswal vs. MIP is 45% when comparing PswalSAL with MIPA, 55% for PswalB1, 55% for PswalB2, and 50% for PswalB3.

Fig. 1
figure 1

Tongue strength values measured by IOPI during different tasks. (TS: tongue strength; MIPa: maximal isometric tongue strength anteriorly, SAL: saliva swallow; B: bolus swallow; Pswal: tongue strength during swallowing)

Full size image

Swallowing Function

All participants completed the entire protocol as described in methods. The procedure was never interrupted for safety concerns.

Concerning swallowing safety, there was no statistically significant difference in PAS score depending on bolus type (χ2(2) = 2.800, p = .247). Figure 2 shows that one participant penetrated without laryngeal clearing on B1 (patient 10), receiving the worst score on PAS-scale. Penetration with laryngeal clearing (score 2 on PAS) was only observed in 1 patient swallowing B2 and in 3 patients swallowing B3.

Fig. 2
figure 2

Number of participants per PAS-score during swallowing different boluses (N = 10). (N: participants obtaining specific PAS-score; PAS: Penetration Aspiration Scale)

Full size image

When analyzing swallowing efficiency, there was no statistically significant difference for the total Pooling score depending on bolus type (χ2(2) = 1.143, p = .565). Visual analysis of Fig. 3 shows no change in the severity of pharyngeal residue depending on bolus type.

Fig. 3
figure 3

Mean Pooling-score during swallowing different boluses (N = 10)

Full size image

Based on these maximum scores, all participants can be classified with moderate (score 8–10) or mild (score 6–8) dysphagia. Table 2 shows the descriptive analysis of the subscales from the Pooling score. No expansion in residue locations has been observed between B1 and B2 or B3. Participant 9 shows more residue when swallowing B2 in comparison with B1 and B3. Management of B2 and B3 seems comparable with management of B1, varying between 1 and 8 swallow-attempts prior to full pharyngeal clearance.

Table 2 Frequency distribution of participants per subscale from the Pooling score (N = 10)
Full size table

Discussion

The use of TMF is widely spread in patients with OD, but associated with malnutrition, a decreased QoL and low adherence rates increasing the risk on aspiration in this vulnerable population [3, 6, 9,10,11,12,13,14]. Since TMF is expected to change sensory input, which is key during swallowing, this unique proof-of-concept study analyzed the impact of adding particles in TMF in patients with OD on tongue strength and swallowing function.

Tongue strength is described as the major contributor to bolus propulsion [21]. Measuring tongue strength during swallowing might quantify the intrabolus pressure needed to perform effective swallows. As far as we know, earlier research concerning PswalB is scarce and mainly comprising non-swallowing tasks in healthy participants [24]. Since tongue strength is described to be decreased in patients with OD [46, 47], examining Pswal in non-healthy participants has an important clinical value. Based on these preliminary findings, particles in TMF does not seem to increase PswalB-values significantly. It is important to note that current standard deviations are high – especially for PswalB2 – and might reflect the heterogeneity which is probably due to this atypical, non-healthy population. Therefore, increasing sample size or including a more homogenous population might differentiate this conclusion. Based on Tobin et al. [27] higher Pswal-values were expected during swallowing the bolus with the biggest particles (PswalB3). Current results do not reflect an increase of mandatory effort when enlarging particle size from 2 mm towards 4 mm. In addition, swallowing boluses with particles requires no more than 55% of the maximal tongue strength; which is not increased in comparison with percentages described in older adults measured when swallowing saliva [48].

A main reason for changing daily diet into TMF and especially pureed food, is enhancing oral bolus formation as well as swallowing safety and efficiency. Contrary to what we expected, the current study shows that adding particles in pureed food does not impact on swallow function. Concerning swallowing safety, aspiration was only once observed when swallowing B1, the bolus without particles. Penetration was observed during swallowing B2, the bolus with small particles, as well as B3, the bolus with bigger particles. Most research about the impact of particle size is limited towards healthy participants and perceptual analyses of their opinion about the easiness of swallowing [49]. Analyses about the impact of TMF on swallowing function is limited towards the effects of bolus viscosity, with a white paper of Newman et al. [50] describing positive effects on swallowing safety, notifying for the possible negative impact on swallowing efficiency. Therefore, comparing the current study with reported literature is difficult.

Since the entire study procedure during FEES was never interrupted for safety reasons, an insight following from this study might be the opportunity of adding a bolus with particles in clinical FEES-assessment beyond the frequently used heterogenous boluses, enabling enhanced patient-centered dietary advices [3, 29].

Based on these preliminary findings, future research can be set up. Increasing sample size may deal with the rather high variability in tongue strength. Variations in particle characteristics as particle size, distribution and hardness are recommended. Expanding FEES examination by means of a videofluoroscopic swallow study would gain further insight in subtle differences of swallowing function, for example the quantification of the amount of residue and visualization of the oral phase [51]. Executing tongue strength measurement analogue with swallowing measurements could enable more detailed analysis about the exact correlation between both parameters, since the technical system allowing for these measurements was not available for the researchers involved in this project. Besides tongue strength, other parameters related to bolus characteristics as the speed of glottal closure, or the impact on respiratory patterns during swallowing, will be interesting to analyze as well and might enlighten more subtle differences between the different boluses. Larger sample sizes combining measurements focusing on patient perception of these boluses and bolus or particle recognition as well as taste and pleasantness are indispensable and necessary to ameliorate patient-centered care in persons with OD on long term. Expanding the amount of boluses to swallow should be interested as well since during this short examination, the impact of fatigue on swallowing function could not be taken into account. Although, fatigue might be a characteristic of the involved participants and is pointed out to predict the risk on dysphagia [52].

Thus, this exploratory study shows no negative effects of adding particles to pureed food tongue strength, swallowing efficiency or safety in patients with OD. Therefore, adding particles to pureed food might create long term opportunities to improve nutrition and food related QoL in patients with OD. This innovative project is the first step in research exploring characteristics of TMF besides bolus volume, viscosity and temperature.

In conclusion, this proof of concept study shows that adding particles to pureed food is feasible, without negatively affecting mandatory effort, swallowing efficiency, or safety in a well-selected group of patients with OD.