The natural swallow: Factors affecting subject choice of bolus volume and pharyngeal swallow parameters in a self-selected swallow

Abstract

Predetermined volumes are used extensively throughout clinical assessment of swallowing physiology, but bolus volumes selected by an individual in their natural swallow can vary greatly from those used in structured assessment. This study aims to identify factors influencing self-selected volume and how the mechanics of self-selected volume swallows differ from predetermined volume swallows. We used pharyngeal high-resolution manometry (HRM) with simultaneous videofluoroscopy to measure swallowing pressures in the velopharynx, hypopharynx, and upper esophageal sphincter (UES). Data were collected from 95 healthy adults during thin liquid swallows of 10 mL and a self-selected comfortable volume. An intraclass correlation coefficient (ICC) was calculated to analyze within-subject self-selected volume reliability. Linear mixed effects regression models were used to examine the association of subject characteristics with self-selected swallow volume and of self-selected volumes on pharyngeal swallowing pressures and timing events. Mean self-selected volume was 16.66 ± 7.70 mL. Increased age (p = 0.002), male sex (p = 0.021), and increased pharyngeal hold area (p = 0.007) were significantly associated with increase in self-selected bolus volume. There was good reliability between subjects’ individual swallow volumes (ICC = 0.80). Velopharyngeal maximum pressure and pressure integral, tongue base duration and maximum pressure, UES pre- and post-swallow maximum pressure, and overall pharyngeal contractile integral decreased significantly with self-selected boluses. Understanding a patient’s natural swallow volume, and how their natural swallow functions, will be important for designing clinical evaluations that place stress on the patient’s natural swallowing mechanics in order to assess for areas of dysfunction.

Introduction

Oropharyngeal swallowing is a complex process involving many sensory input sources, the coordination of multiple paired muscles, and the movement of aero-digestive tract structures to propel a bolus from the oral cavity through the upper esophageal sphincter (UES) while protecting the airway. Bolus characteristics such as volume [1,2,3,4,5,6], viscosity [1,2,3], taste [7, 8], temperature [9, 10], and bolus delivery mechanism [11] each modulate the physiologic properties of swallowing including timing events, muscle movement, and pressure generation. An understanding of normal swallowing mechanics is key to evaluation and treatment of patients with disordered swallowing, but natural conditions are rarely applied to swallow evaluation in the clinical setting.

It has been demonstrated that bolus volume contributes to the mechanics of pharyngeal swallowing [1,2,3,4,5,6]. While different methods of measurement have yielded varying results, studies utilizing high-resolution manometry (HRM) to evaluate the pharyngeal swallow in normal subjects have demonstrated significant effect of bolus volume on velopharyngeal pressure duration, maximum tongue base pressure, tongue base pressure rise rate, UES opening duration, UES opening pressure, and total swallow duration [4,5,6]. Hoffman et al. [4] found that velopharyngeal pressure duration, UES opening duration, and total swallow duration increased with increasing bolus volume, while maximum tongue base pressure decreased with increasing bolus volume. An increase in UES opening duration and pressure with increasing bolus volume was also found by Lin et al. [5] and Ryu et al. [6], whereas Ryu et al. [6] found decreases in UES pre-opening and post-opening pressure with increasing bolus volume. In these studies, boluses were delivered at predetermined volumes ranging from 2 to 20 mL. While this is an externally standardized method used extensively throughout research and clinical assessments of swallow, there is concern that the individual variability of an unmeasured swallow volume could impact the interpretation of this standardized data. Bolus volumes selected during a natural swallow can vary significantly [12,13,14] from those used during structured assessment. Since external factors exert so much influence on swallowing function, it is reasonable to consider whether the structured swallow assessment leads to significant difference in pharyngeal function compared to a natural swallow.

Prior studies have investigated the volumes typical of a natural swallow, which is defined as a swallow in which subjects are not given a predetermined volume to drink, but choose a swallow volume based on what is most comfortable to swallow in full, without piecemeal deglutition. The natural swallow volume will be referred to as the self-selected volume swallow throughout the remainder of this paper. Reported volumes range from 12.75 to 25 mL [15,16,17], with increased volumes in male subjects [12,13,14]. While methods of measurement varied, these studies indicated that self-selected bolus volumes can vary greatly between studies and from volumes typically used in research and clinical assessments of swallowing. Age-related differences in self-selected volume have varied. Some studies identified no age-related volume change [13], but others found that younger subjects took larger volume sips [15]. Lawless et al. [13] found that greater subject height was correlated with greater swallow volume, suggesting that sip size may be related to overall body size.

When swallowing a predetermined volume, the individual must adapt their swallow to accommodate for the properties of the bolus. The brainstem integrates the many factors affecting swallow mechanics to produce the required amount of force, in the appropriate sequence, to drive the bolus from the oral cavity through the UES. The most efficient swallow will utilize the lowest amount of pressure required to complete this action. It is possible that when an individual is allowed to select their most comfortable bolus volume, this size is determined using input from the brainstem level to create a swallow requiring the least amount of pressure. A more complete understanding of natural swallowing conditions and mechanics can inform the understanding of alterations in swallowing and aid in identification of swallow impairment.

This study aimed to address the following questions:

1. Is the average self-selected bolus volume within our subject population consistent with previously reported values?

2. How do demographic and clinical characteristics (age, sex, pharyngeal size) contribute to self-selected bolus volume?

3. How variable is the self-selected volume within subjects?

4. Does use of a self-selected bolus volume lead to changes in pharyngeal swallowing parameters measured by high-resolution pharyngeal manometry?

We hypothesized that (1) self-selected volume would increase with age; (2) self-selected volume would differ by sex, predictable by differences in pharyngeal size; (3) within-subject swallow volume variability will be low; and (4) if volumes were considered equal, pharyngeal pressure generation would be less with self-selected swallows, indicating an individual’s most efficient swallow.

Materials and Methods

Participants

Data were obtained from a normative database of pharyngeal swallowing pressures comprised of data collected from 101 healthy adults. There were 95 participants (41 male, 54 female) with complete datasets, ranging from ages 21 to 89 years (mean 51.8 years). There were 48 subjects younger than 50 years of age and 47 subjects having 50 years of age or older. All individuals were without known neurologic, respiratory, or gastrointestinal disorders and denied having any symptoms of dysphagia. All participants provided written informed consent and all data were collected under research protocols approved by the University of Wisconsin-Madison Health Sciences Institutional Review Board (2013–1227).

Equipment

The ManoScan ESO system (Medtronic, Minneapolis, MN) was used for pharyngeal pressure data collection in conjunction with simultaneous videofluoroscopy. A pressure catheter with 36 sensors spaced 1 cm apart and an outer diameter of 2.75 mm was placed through the nostril after application of topical viscous lidocaine hydrochloride such that 9 to 15 sensors were located in the pharynx and UES region, depending on the height of the individual. Participants were given a period of approximately 6 min of rest to acclimate to catheter placement prior to delivery of test boluses. The system registers contact pressures between – 20 mmHg and 300 mmHg with a 2 mmHg resolution and a 50 Hz sampling rate. Each sensor is 0.4 mm long and collects circumferential input, which is averaged, and the mean pressure is recorded as the pressure for the individual sensor. Catheter calibration was performed before each subject in accordance with the manufacturer’s specifications.

Continuous videofluoroscopy was captured in the lateral plane ((OEC 9900, General Electric, Fairfield, CT). The video was digitized and recorded at 30 frames per second on a DVD + RW (DVO-1000MD, Sony, Park Ridge, NJ) for offline analysis. The videofluoroscopic frame was adjusted to include the incisors, cervical vertebrae, nasal border of the soft palate, and the cervical esophagus.

Data Collection

During simultaneous High-Resolution Manometry (HRM) and videofluoroscopy, participants swallowed 10 trials of 10 mL and up to 10 sips of a participant-selected (“self-selected”) comfortable volume of thin liquid barium sulfate (40% w/v; Varibar, Bracco Diagnostics, Monroe Township, NJ). All participants were situated upright with the head in a neutral position. The premeasured 10 mL boluses were delivered to the oral cavity via syringe, with the tip placed posterior to the teeth. Participants were given a cue to swallow, with 10 trials per subject. For the self-selected comfortable volume task, participants were given a 12 oz paper cup and straw with 100 mL of thin liquid barium and asked to take a sip of their normal, comfortable volume, which they could swallow without use of piecemeal deglutition. Participants were required to use the straw due to the difficulty of taking a cup sip with the HRM catheter in the nose. Each subject took repeated sips until the 100 mL total was completed, or up to 10 total sips to match the premeasured volume trial number and limit radiation exposure. The liquid-containing cup was weighed before and after each swallow to determine the volume taken in each sip.

Data Analysis

Self-selected Bolus Volume

For each subject, the last self-selected swallow trial was excluded in order to account for the last volume consisting of just the remaining liquid, rather than a true natural swallow volume.

Pharyngeal Pressures

A custom, automated analysis program was used to extract HRM data and analyze pharyngeal pressure parameters for regions of interest [18, 19]. Regions identified were the velopharynx, tongue base, hypopharynx, and UES (Fig. 1). These regions were identified manometrically as detailed in prior studies [19, 20]. Briefly, the velopharynx is the superior most region of pharyngeal swallowing pressure, identifiable by bimodal or peak-plateau waveform shape. Directly caudal is the tongue base, which has pressure contributions from the base of tongue and superior and middle pharyngeal constrictors. The hypopharynx is identified by multi-peak pressure waves created by contributions from the middle and inferior constrictors as well as the laryngeal structures. The upper esophageal sphincter has muscular contributions from the inferior pharyngeal constrictors, cricopharyngeus, and striated cervical esophagus musculature and demonstrates a decrease in resting pressure during swallowing [21]. Swallowing parameters measured and recorded for each swallowing trial included pressure maximum, pressure duration above baseline, and pressure integral for the velopharynx, hypopharynx, and UES. Additionally, pre-relaxation maximum pressure, post-relaxation maximum pressure, minimum pressure, and duration of minimum pressure were captured for the UES. Overall swallow duration, defined as the time elapsed from onset of velopharyngeal pressure rise to post-swallow UES pressure peak, and pharyngeal contractile integral were also recorded. Pharyngeal swallow pressure and timing data were averaged across all swallows for both the 10 mL and self-selected condition.

Figure 1.

Simultaneous videofluoroscopy and pharyngeal High-Resolution Manometry (HRM). a Pharyngeal swallow pressure spatiotemporal plot. b Pressure waveforms that are represented in the pressure spatiotemporal plot in (a) separated into velopharynx, tongue base, hypopharynx, and upper esophageal sphincter regions. c Videofluoroscopic still image at time of Pharyngeal hold, before the swallow shown in (a) and (b), with pharyngeal hold area (PaH) outlined.

C2-C4 Height

Vertebral C2-C4 heights were measured as a proxy for height [22]. The C2-C4 measurement was made using an imported videofluoroscopic frame in ImageJ (National Institutes of Health, Bethesda, MD). A pixel to centimeter conversion was made using the distance between the beginning of one manometric sensor and the next as a known scalar of 1 cm. C2–C4 distance was measured in pixels between the anterior inferior corner of the C2 vertebrae and the anterior inferior corner C4 vertebrae using the measuring tool in ImageJ.

Pharyngeal Areas

The area of the pharynx in the resting position (Pharyngeal hold area, PaH) was measured using imported videofluoroscopic frames in ImageJ. The PaH is the measure of the resting, empty pharynx in the lateral view hold frame with the bolus held in the oral cavity, just before the bolus makes any posterior movement [23, 24] (Fig. 1). The resting pharyngeal space was outlined with a posterior border along the posterior pharyngeal wall from the mid-portion of the atlas tubercle down to the level of the top of the arytenoid cartilages. Inferiorly, the border was carried anteriorly over the arytenoid cartilages and forward to outline the epiglottis. The line was continued superiorly to outline the anterior pharyngeal border over the vallecula, tongue base, and pharyngeal portion of the soft palate to the level of the posterior nasal spine. The superior border was created with a straight line from the posterior nasal spine to the midpoint of the tubercle [24, 25]. The pixel to centimeter calibration used above was used to calculate the outlined area in ImageJ in cm2. PaH has been shown to vary by sex, though not height or weight24.

Statistical Analysis

Descriptive sample statistics of self-selected bolus volume, C2-C4 height, and PaH were calculated for all subjects, by sex and by age group (age less than 50 or age greater than or equal to 50) based on an overall subject mean age of 51.8 years. Sample statistics for volumes, swallow pressures, and swallow timing events were determined by taking the mean of all an individual subject’s trials, then calculating the statistic of interest across all subjects. Intraclass correlation coefficients (ICCs) and their associated 95% Confidence Intervals (CI), along with box-and-whisker plots, were used to analyze within-subject self-selected volume reliability overall, by gender, and by age group. ICCs were computed using the SAS %ICC9 macro with absolute agreement as the reliability estimate [26]. A dichotomous age group variable was used (rather than continuous age) for descriptive statistics and ICC analysis based on subject population mean for simpler categorical interpretation.

In regression models, all swallow trials were used for analysis except for the final self-selected swallow, as stated above. Linear mixed effects regression models with a repeated subject statement and a compound symmetry covariance structure were used to examine the association of subject characteristics with self-selected swallow volume. Age was analyzed as a continuous variable. Similarly, a mixed effects model with repeated subject statement was used to examine the effect of trial number on self-selected swallow volume. Each participant’s total number of trials was accounted for in the model.

Differences in swallowing pressure and timing events between self-selected bolus swallows and a predetermined bolus of 10 mL were also examined using linear mixed effects regression models. The models included a repeated subject statement and accounted for swallow volume.

Descriptive analysis was performed using SPSS, version 26 (SPSS, Inc, an IBM Company, Chicago, IL) and regression modeling was implemented using SAS software (version 9.4, SAS Institute Inc., Cary, NC). No statistical corrections were used; p-values less than 0.05 were considered statistically significant.

Results

Clinical Characteristics

The mean self-selected swallow volume for all subjects was 16.66 mL ± 7.70. Male and female subjects had mean self-selected volumes of 18.82 mL ± 8.95 and 15.02 mL ± 6.19 (Table 1). Descriptive sample statistics for self-selected swallow volumes, C2-C4 height, and pharyngeal hold area are summarized in Table 1. In linear mixed effects models, increased age, male sex, and increased pharyngeal hold area were significantly associated with an increase in self-selected swallow volume (p = 0.002, p = 0.021, and p = 0.007, respectively) (Table 2). There was no statistically significant difference in self-selected volume for varying C2-C4 heights (p = 0.539). When adjusting for “size of system” variables (C2–C4 height and pharyngeal hold area) using a multivariable analysis, subjects with male sex still had a greater self-selected swallow volume than female subjects (p = 0.049) (Table 2).

Table 1.

Mean self-selected volumes, C2-C4 height, and pharyngeal area.

Group	Self-selected bolus volumes			C2-C4 height			Pharyngeal hold area
	n	Mean (ml)	SD	n	Mean (cm)	SD	n	Mean (cm2)	SD
All patients	95	16.66	7.70	92	3.54	0.39	91	10.69	2.67
Age
<50	48	14.32	6.91	47	3.49	0.38	47	10.6974	2.99
50+	47	19.06	7.80	45	3.60	0.40	44	10.69	2.32
Sex
Male	41	18.82	8.95	41	3.76	0.35	40	10.94	2.64
Female	54	15.02	6.19	51	3.37	0.33	51	10.50	2.71

SD = standard deviation

Table 2.

Results of linear mixed effects regression to examine the association of subject characteristic with self-selected bolus volume. A multivariate analysis was included to analyze whether individual effect of male sex could be explained by adjusting for “size of system” variables, C2-C4 height, and pharyngeal hold area. Significant results (p < 0.05) in bold. CI = confidence interval.

Bivariate analysis
Variable	Fixed effects ß	Standard error	t value	95% CI	p value
Age	0.11	0.03	3.12	[0.04, 0.20]	0.002*
Sex	3.64	1.55	2.35	[0.56, 6.71]	0.021
C2-C4 height	1.26	2.05	0.62	[− 2.81, 5.33]	0.539
Pharyngeal hold area	0.72	0.26	2.77	[0.20, 1.24]	0.007
Multivariable analysis
Variable	Fixed effects ß	Standard error	t value	95% CI	p value
Sex	3.23	1.62	2.00	[0.02, 6.44]	0.049
C2-C4 height	− 1.52	2.06	− 0.74	[− 5.61, 2.58]	0.463
C2-C4 height	− 1.52	2.06	− 0.74	[− 5.61, 2.58]	0.463
Pharyngeal hold area	0.70	0.26	2.69	[0.18, 1.21]	0.009

Swallow Volume Variability

Box plots demonstrating intra-subject self-selected swallow volume variability are shown in Fig. 2 to represent how the volume chosen differed between a subject’s individual sip trials. For all subjects, the intraclass correlation coefficient (ICC) was 0.80, 95% CI [0.73, 0.84]. This was calculated as a quantitative representation of correlation between a subject’s self-selected swallow volumes. A value of 0.80 indicates good reliability between an individual’s choices of swallow volume. ICC values and corresponding coefficients of variation (CoV) are shown in Table 3. When grouping by sex, females’ swallow volume was slightly more reliable than males’ (0.82, 95% CI [0.75, 0.88] vs 0.74, 95% CI [0.63, 0.83], respectively), but the difference was not significant due to overlapping confidence intervals. When grouping by age, subjects less than 50 years had less variability (ICC = 0.85, 95% CI [0.78, 0.90]) between swallows than older subjects (ICC = 0.70, 95% CI [0.58, 0.80]), but there was not a statistically significant difference between the two groups. When an individual subject’s total number of swallow trials was adjusted for in a mixed effects model, it was demonstrated that for every one unit increase in trial number, the average expected swallow volume decreased by 0.18 mL (p = 0.037).

Figure 2.

Within-subject self-selected swallow volume variation demonstrated via boxplot and intraclass correlation coefficient (ICC) for a) all subjects; b) female subjects; c) male subjects; d) age < 50; and e) age 50+.

Table 3.

Intra-subject self-selected volume variability for all subjects by sex and by age group using Intraclass correlation coefficients and coefficients of variation. ICC = intraclass correlation coefficient; CoV = coefficient of variation; CI = confidence interval.

	ICC	95% CI	CoV	95% CI
All subjects	0.79	[0.73, 0.84]	0.23	[0.20, 0.25]
Sex
Female	0.82	[0.75,0.88]	0.19	[0.16, 0.22]
Male	0.74	[0.63,0.83]	0.26	[0.22, 0.31]
Age
<50	0.85	[0.78,0.90]	0.2	[0.17, 0.24]
50+	0.7	[0.58,0.80]	0.25	[0.21, 0.29]

Pharyngeal Swallow Pressure and Timing

Using High-Resolution Manometry, pharyngeal swallowing pressure and timing events were recorded for both 10 mL volume swallows and self-selected volume swallows. A summary of mean recorded swallowing pressure and timing parameters for both the 10 mL swallow and self-selected swallow conditions is shown in Table 4. The results of linear mixed effects models examining the difference in pressure and timing events between the two conditions, when swallow volume has been adjusted for, are also reported in Table 4. In these models, fixed effects parameter estimates represent the change in outcome we would expect to see in self-selected swallows compared to predetermined swallows if swallow volume was held constant. Results of the linear effects model examining regional events are reported by region below.

Table 4.

Mean swallowing pressures and timing events for 10 mL and self-selected swallows as measured by High-Resolution Manometry and linear mixed effects model results. Descriptive statistics do not account for varying swallow volumes and were calculated by averaging all trials for each subject and then computing sample statistics across all subjects. Expected change in outcome between self-selected swallow and 10 cc swallow, when adjusting for swallow volume, are shown in column “Fixed effects β*” along with corresponding p value. For all models, the reference condition is predetermined syringe-delivered boluses and swallow volume was adjusted for. Significant results (p < 0.05) in bold. SD = standard deviation; UES = upper esophageal sphincter; P_max = maximal regional pressure; Integral = cumulative regional pressure; P_max pre = maximum pre-opening pressure; P_max post = maximum post-closure pressure; Nadir duration = duration of UES opening; CI = confidence interval.

	10ml		Self		Linear mixed effects model
	Mean	SD	Mean	SD	Fixed effects β	95% CI	p value
Velopharynx
Duration (s)	0.76	0.18	0.78	0.21	− 0.02	[− 0.03, 0.00]	0.120
Pmax (mmHg)	164.06	57.69	164.29	55.57	− 5.87	[− 8.43, − 3.32]	< 0.0001
Integral (mmHg*s)	120.16	61.89	127.32	64.82	− 10.19	[− 14.00, − 6.38]	< 0.0001
Tongue Base
Duration (s)	0.49	0.13	0.46	0.13	− 0.018	[− 0.03, − 0.01]	0.003
Pmax (mmHg)	123.59	42.82	120.84	38.42	− 3.21	[− 5.39, − 1.02]	0.005
Integral (mmHg*s)	77.33	39.78	75.85	40.51	0.72	[− 1.69, 3.14]	0.553
Hypopharynx
Duration (s)	0.38	0.12	0.38	0.13	− 0.00	[− 0.01, 0.01]	0.966
Pmax (mmHg)	151.51	53.08	151.87	55.53	− 1.91	[− 5.02, 1.20]	0.225
Integral (mmHg*s)	47.22	28.62	48.56	30.85	0.56	[− 1.05, 2.16]	0.493
UES
Pmax pre (mmHg)	108.12	43.25	100.35	43.21	− 13.48	[− 17.40, − 9.56]	< 0.0001
Pmax post (mmHg)	212.63	63.15	213.69	63.13	− 4.55	[− 7.98, − 1.13]	0.010
Pmax post (mmHg)	212.63	63.15	213.69	63.13	− 4.55	[− 7.98, − 1.13]	0.010
Nadir durationg (s)	0.49	0.14	0.50	0.14	0.00	[− 0.02, 0.02]	0.964
Pnadir (mmHg)	4.47	6.92	5.14	7.99	− 0.25	[− 0.89, 0.38]	0.428
Overall
Duration (s)	0.82	0.15	0.85	0.18	− 0.02	[− 0.03, 0.00]	0.077
Integral (mmHg*s)	244.72	100.76	251.69	109.53	− 8.85	[− 13.32, − 4.37]	0.0002

Velopharynx

For the velopharynx, based on the linear effects model, we would expect to see, on average, a 5.87 mmHg decrease in maximum pressure in self-selected bolus swallows compared to the predetermined 10 mL boluses, when volume was accounted for (p < 0.0001). There would also be an expected decrease of 10.19 mmHg*s in pressure integral for the self-selected swallow (p < 0.0001).

Tongue Base

In the tongue base, self-selected swallows demonstrated a small but significant expected decrease in pressure duration compared to the 10 mL swallows (p = 0.003). The maximum tongue base pressure was also less for the self-selected swallows (p = 0.005).

Hypopharynx

For the hypopharynx, there were no statistically significant differences in pressure duration, maximum pressure, or pressure integral between the self-selected swallows and the 10 mL swallows when volume was accounted for in the linear effects model.

Upper Esophageal Sphincter

For the UES, based on the linear effects model, self-selected swallows demonstrated average expected decreases of 13.48 mmHg in maximum pre-relaxation pressure (p < 0.0001) and 4.55 mmHg in maximum post-relaxation pressure (p = 0.010). There were no statistically significant differences in UES nadir duration or UES nadir pressure between self-selected bolus swallows and 10 mL bolus swallows.

Overall

The overall swallow duration was not significantly different between the self-selected and 10 mL swallows. The pharyngeal contractile integral was on average 8.85 mmHg*s less for the self-selected bolus swallows compared to the 10 mL bolus swallows (p = 0.0002).

Discussion

This study aimed to characterize natural swallowing conditions and mechanics by analyzing factors that contribute to self-selected swallow volume, how self-selected volume varies between an individual’s swallows, and how pharyngeal swallow pressure and timing parameters differ between swallows of self-selected boluses and predetermined syringe-delivered boluses.

Clinical Characteristics

The mean self-selected volume for all subjects was consistent with some existing studies [15], but less than others [12,13,14]. Volumes measured in this study were using thin liquid barium and the smaller volumes noted can be interpreted in this context. Adnerhill et al. [12] demonstrated that barium swallows tend to be smaller than other thin liquid swallows which is consistent with other studies which measured larger self-selected volumes using water [13, 14]. Age, sex, and pharyngeal hold area were associated with subject volume choice with increased self-selected volumes seen for male subjects, older subjects, and with increasing pharyngeal hold area. Male subjects had larger self-selected volumes than female subjects, even when accounting for the size of system variables, C2–C4 height, and pharyngeal hold area, which is also consistent with previous studies [12, 13, 17]. For all subjects, increasing pharyngeal hold area was associated with increased self-selected volumes. Likely, when there is a larger pharyngeal space to fill, higher bolus volumes are required to generate the same amount of pressure. Previous studies [2, 27] have shown that pharyngeal swallowing pressures and UES nadir pressures increase with age. In the setting of increased UES nadir, higher driving pressures are required to propel a bolus through the UES. Larger bolus volumes are associated with increased velopharyngeal pressure duration, tongue base maximum pressure, and UES opening duration [4]; therefore, it follows that for older subjects who have higher UES opening pressures, larger volume swallows can increase sensory input and pharyngeal pressure generation to allow for bolus movement through the UES. These factors inform natural volume choice, potentially leading to the demonstrated increased self-selected volumes in older individuals.

Swallow Volume Variability

Overall, there was low self-selected volume variability between individual subject’s sip trials, though this variability increased slightly with age when subgroup analysis was performed. Variation in how a repeated task is performed is expected [21, 28]. Stable motor systems have the optimal amount of motor variability to adequately complete the task while adapting for different conditions. Traditionally, motor variability is thought of as the result of errors in prediction of parameters required for motor task completion. When the task is optimized through repetition, variability decreases as errors are eliminated [29]. Our hypothesis that subjects would have low variability between self-selected volume trials was supported by an ICC approaching 1, indicating good correlation between an individual’s choices of swallow volumes. When given a choice in volume, the natural response is to select a similar volume each time. This potentially demonstrates that the self-selected swallow is part of a stable motor task, allowing for efficient sensory integration, efficient task completion, and reduction of motor error. It is well established that motor variability increases with age [21] which supports our findings that within-subject bolus volume variability increases slightly for older subjects.

Pharyngeal Swallow Pressure and Timing

It is known that bolus volume [1,2,3,4,5,6], viscosity [1,2,3], taste [7, 8], temperature [9, 10], and delivery mechanism [11] are associated with changes in swallowing mechanics. When determining whether there were differences between self-selected and predetermined volume swallows, it was, therefore, important to control for other potentially influential factors. Differences due to bolus volume and bolus delivery mechanism were controlled for through statistical analysis and by asking participants to hold the bolus in their oral cavity until cued to swallow. Therefore, objective assessment of differences in swallowing mechanics due to self-selected versus predetermined volume can be more clearly undertaken.

During the pharyngeal swallow, when the bolus reaches the tongue base, the posterior and lateral pharyngeal walls move inward and the tongue base retracts posteriorly to meet the pharyngeal walls, generating driving force at the bolus tail. Tongue base movement is considered the major pressure-generating force for bolus propulsion through the pharynx [30]. It appears that during self-selected swallows, lower driving forces in the manometric tongue base region are required, for less time. This region consists of pressures generated circumferentially from the tongue base and pharyngeal walls, with greater pressures generated posteriorly [31]. Further, when a self-selected swallow volume was used, there was no change in the UES relaxation pressure or opening duration. The UES was relaxed for the same amount of time at the same pressure for the self-selected and predetermined swallows, but a lower maximum tongue base and pharyngeal wall driving pressure was required for the self-selected swallow. High-resolution manometry catheters used in this study collect circumferential data and record an average of all sensors. Lower maximum pressure values could reflect lower, but more uniform, circumferential pressures that are able to provide a consistent driving force on all sides of the bolus, thus more efficiently causing bolus movement. There were no statistically significant differences detected in tongue base pressure integrals between the self-selected and predetermined swallows. It is unclear if this result is due to inadequate power to detect small differences or truly similar results, but lack of difference could indicate that equal driving forces were generated, but with lower maximum pressures. This suggests that subject-selected swallow allows optimization of swallowing events creating an efficient, low-pressure swallow.

Prior work has demonstrated that velopharyngeal and UES pressure waveform variability is greater for self-selected swallows compared to predetermined, 10 mL syringe-delivered bolus swallows [21]. For the subject-driven task, the UES responds to the slight variation in volume between tasks by producing a larger range of normal swallowing pressures. The present study builds on this finding by showing that, in addition to pressure variation within the self-selected task, swallowing pressures can differ between self-selected swallows and predetermined swallows, emphasizing the importance of assessing a patient’s swallow in the setting most analogous to their natural swallow.

Limitations

There are several limitations to this study. First, while this analysis attempted to compare predetermined volume swallows to self-selected, or natural, swallows, data were collected in a structured environment. Subjects were instructed to take sips in a comfortable manner, but as it is generally not a natural task to take repeated swallows, results may not be fully generalizable to natural day-to-day swallowing behavior. Second, it is known that bolus volume affects pharyngeal pressure and timing events. In order to allow subjects to select a natural swallow, it was not possible to control for the effects of volume in the experimental setting. Differences in volume between the self-selected swallows and 10 mL swallows were accounted for during statistical analyses. Discrepancies in interpretation between descriptive sample statistics and regression results as shown in Table 4 are likely due to differences in volume between the 10 mL swallow and each subject’s self-selected volume, which may have been more or less than 10 mL. Future studies may be able to control for differences due to volume more directly by providing a predetermined bolus of volume equal to a subjects self-selected volume. Third, it is known that bolus delivery mechanism also affects swallowing mechanics, coordination, and functional outcome. To control for differences in bolus delivery mechanism between self-selected and predetermined swallows, it was necessary to require subjects to hold the bolus, delivered either by syringe or cup and straw, in their oral cavity and only swallow once cued. This allowed for more direct assessment of differences due to self-selected versus predetermined volume, but there is some limitation in the generalizability of interpretation since we were not able to fully control for bolus delivery mechanism.

Future Directions

This study evaluated clinical and demographic predictors of the self-selected swallow volume, volume choice variability, and self-selected bolus effect on pharyngeal swallow pressures and timing in a normal population. Future studies will build on the insights provided for a normal population to determine if these parameters may change in cases of pathology such as Parkinson disease, stroke, or head and neck cancer.

Conclusion

Self-selected swallow volume is associated with age, pharyngeal hold area, and male sex. Individual’s swallow volumes were highly correlated, suggesting that each individual has an intrinsic comfortable sip volume, potentially established at the level of the brainstem. Natural swallows were associated with lower maximum tongue base driving pressures, but similar cumulative pressures, suggesting that a natural swallow utilizes more uniform, efficient swallowing dynamics than structured swallowing. These findings contribute to a more robust understanding of natural swallowing dynamics and how integration of external factors contribute to motor control. This analysis is clinically relevant because swallow evaluations are often very structured, which may not offer a full picture of how the patient is swallowing outside of the clinical setting. Understanding a patient’s natural swallow volume, and how their natural swallow functions, will be important for designing clinical evaluations that place stress on the patient’s natural swallowing mechanics in order to assess for areas of dysfunction.