Maximum lingual pressure impacts both swallowing safety and efficiency in individuals with amyotrophic lateral sclerosis

Abstract

Background:

Although reduced lingual strength is a confirmed early manifestation of amyotrophic lateral sclerosis (ALS), its functional impact on swallowing remains unclear. We therefore sought to examine relationships between maximum anterior isometric lingual pressure (MAIP) with swallowing safety, swallowing efficiency, and swallowing timing metrics in a large cohort of individuals with ALS.

Methods:

Ninety-seven participants with ALS completed a standardized videofluoroscopic swallowing examination (VF) and lingual pressure testing (Iowa Oral Performance Instrument). Duplicate and blinded ratings of the Penetration-Aspiration Scale (PAS) and Analysis of Swallowing Physiology: Events, Kinematics and Timing (ASPEKT) percent efficiency (%C2–C4²) and timing (laryngeal vestibule closure (LVC) duration: amount of time (milliseconds, msec) between LVC onset and laryngeal vestibule opening; time-to-LVC: hyoid burst to onset of LVC (msec); and swallow reaction time: interval between bolus passing ramus of mandible and onset of LVC (msec)) were performed across bolus trials. Swallowing safety (safe PAS: 1, 2, 4; unsafe PAS: 3, 5, 6, 7, and 8) and efficiency (inefficient: ≥3% worst total residue) were derived. Statistical analyses including descriptives, binary logistic regressions, and Spearman’s rho correlations were performed (α = 0.05).

Key Results:

Mean MAIP was 36.3 kPa (SD: 18.7). Mean MAIP was higher in those with safe swallowing as compared to those who penetrated (mean difference: 12 kPa) or aspirated (mean difference: 18 kPa). Individuals with efficient swallowing demonstrated higher MAIP than those with inefficient swallowing (mean difference: 11 kPa). Binary logistic regression analyses revealed increasing MAIP was significantly associated with a 1.06 (95% CI: 1.03–1.09) and 1.04 (95% CI: 1.01–1.06) greater odds of safe and efficient swallowing, respectively. No relationships were observed between MAIP and swallow reaction time across all bolus trials. Longer time-to-LVC (5 ml thin liquid: r_s = −0.35, p = 0.002; cup sip thin liquid: r_s = −0.26, p = 0.02; moderately thick liquid: r_s = −0.28, p = 0.01) and prolonged LVC duration (cup sip thin liquid, r_s = −0.34, p = 0.003) were associated with lower MAIP.

Conclusions and Inferences:

Reduced lingual strength was confirmed in this group of 97 individuals with ALS that was associated with a diminished ability to effectively transport boluses and aide in laryngeal vestibule closure to prevent entry of material into the airway.

1 |. INTRODUCTION

Swallowing food and liquids is an important homeostatic function required to maintain adequate nutrition and hydration.¹ Deglutition includes two functional components: the ability to avoid airway invasion (i.e., penetration and/or aspiration) during swallowing (i.e., safety),² and efficient bolus transportation and clearance (i.e., efficiency).^3,4 The tongue is a functional muscular hydrostat,⁵ comprising eight interdigitating intrinsic and extrinsic muscles that work in concert to propel and clear a bolus from the oral cavity into the pharynx during the oral phase of swallowing (swallowing efficiency)³ and that aides in laryngeal vestibule closure (LVC) during the pharyngeal phase via retraction of the tongue base to the pharyngeal wall (swallowing safety).⁶ In amyotrophic lateral sclerosis (ALS), the tongue is particularly susceptible to degeneration and recognized as an early orofacial structure to undergo anatomic, morphologic, and physiologic decline.^7,8 Documented changes include reductions in myelinated lingual fibers,⁹ lingual atrophy,¹⁰ the tongue noted to become more rectangular,¹¹ and reduced lingual thickness.¹² These underlying structural changes give rise to physiologic reductions in lingual strength in individuals with ALS.^13–15 In turn, it is presumed that the reduced maximum anterior isometric lingual pressure (MAIP) affects swallowing function; however, investigations in this area are limited.

One aspect of lingual function, pressure generation, can be assessed and is of particular interest and relevance to the study of individuals with ALS given the aforementioned degradation of the lingual musculature during disease progression.¹⁶ Lingual pressure testing has been used in several patient populations to examine relationships between lingual pressure and swallowing function.^17–20 Regarding swallowing safety, this research has primarily revealed that a reduced ability to generate MAIP is associated with aspiration risk in the healthy elderly,²⁰ the frail elderly,²¹ and stroke patients.²² Further, MAIP <40 kilopascals (kPa) represents an identified functional threshold for increased aspiration risk.²⁰ In addition, Steele and Cichero noted an increased susceptibility to aspiration in individuals with reduced control over timing of tongue to palate pressure release.²³ Taken together, this suggests that both lingual pressure generation and timing of lingual pressure generation are important contributors to maintaining swallowing safety.

Given the tongue is considered the “primary pressure driver” of swallowing,²⁴ establishing the relative utility and role of lingual pressure testing as it relates to impairments in swallowing efficiency has also been an area of interest. Research from several patient populations has noted associations between reduced lingual pressure and oral phase swallowing impairments that include impaired bolus manipulation, reduced tongue to palate pressure, prolonged oral transit times, and increased mealtime duration.^18,19,22 Reduced lingual pressure generation and tongue driving force are also linked to increased post-swallow residue²⁵ which may increase susceptibility to aspiration on ensuing swallows.^26–28

Although reduced maximum tongue strength is documented in ALS,^7,13,14,29 there is a paucity of data examining relationships between lingual pressure in ALS and swallowing safety and efficiency. Further, how changes to lingual pressure during ALS disease progression may relate to changes in physiologic mechanisms that underlie swallowing function such as the timing and duration of laryngeal vestibule closure has not been investigated. We therefore aimed to determine MAIP profiles in persons with ALS and compare these to established healthy reference data. We also aimed to examine the relationship between MAIP with swallowing safety, swallowing efficiency, and the temporal aspects of swallowing function in individuals with ALS. We hypothesized that (1) secondary to previous findings of early and extensive changes to the tongue during ALS disease progression,^7,8,30,31 individuals with ALS would demonstrate reduced MAIP, compared to healthy reference data; (2) given associations between reduced MAIP and impairments in swallowing safety and efficiency in other patient populations,^17,19,23,32 unsafe and inefficient swallowing would be associated with lower MAIP in ALS; (3) as compared to swallowing safety, swallowing efficiency would be more sensitive to and affected by reductions in MAIP due to the tongue’s role as the primary pressure driver for bolus transport during deglutition; and (4) reductions in lingual pressure may be associated with other physiologic changes to the bulbar musculature impacting timely swallowing movements and, thus, would be associated with prolonged time-to-LVC and a longer swallow reaction time.

2 |. MATERIALS AND METHODS

2.1 |. Participants

Ninety-seven individuals were enrolled in this study. Participants were recruited from an academic neurology multidisciplinary clinic. Inclusion criteria included (1) a confirmed diagnosis of ALS (El-Escorial criteria revisited)³³ made by neuromuscular neurology specialist; (2) participant was not feeding tube dependent (still consuming some form of oral intake); (3) no allergies to barium; (4) no history of stroke, head and neck cancer, or other disorders or surgeries (e.g., head and neck surgery) impacting swallowing; and (5) not pregnant. This current study was a subset of a larger longitudinal natural history wherein participants attended a comprehensive research evaluation every 3 months. The present study includes participant data from the baseline evaluation. This study was approved by the university Institutional Review Board and conducted in accordance with the Declaration of Helsinki. Each participant provided written informed consent upon enrollment.

2.2 |. Procedures

2.2.1 |. Lingual pressure testing

Maximum anterior isometric lingual pressure (MAIP) was assessed using a handheld lingual pressure transducer, the Iowa Oral Performance Instrument (IOPI; IOPI Medical LLC). The IOPI device was set in “peak” mode during testing, and the air-filled bulb was placed in the participant’s mouth on the hard palate directly behind the front teeth (i.e., anterior placement position). Participants were instructed to press the front of their tongue against the bulb as hard as possible. The amount of lingual pressure generated was expressed on the LCD screen of the IOPI device in kilopascals (kPa). Three attempts were performed with a one-minute rest period between each trial, with additional rest time provided if the participant requested more time to recover. The highest (maximum) of the three trials was used in subsequent data analysis.

2.2.2 |. Videofluoroscopic swallowing examination

Swallowing function was assessed using the gold standard videofluoroscopic swallowing evaluation (VF). Videofluoroscopic procedures and analyses were performed in accordance with previously described methods.³⁴ Participants were seated upright at 90 degrees in a TransMotion Medical TMM3 Videofluoroscopy Swallowing Study Treatment Chair. Participants were imaged in the lateral plane using a properly collimated Phillips BV Endura System fluoroscopic C-arm unit (GE OEC 8800 Digital Mobile C-Arm System). The VF was performed using continuous fluoroscopy, and a TIMS DICOM system (Version 3.2, TIMS Medical) recorded high-resolution images at a rate of 30 frames per second. Images were automatically spliced into individual bolus trials for data analysis. A standardized bolus presentation was administered consisting of the following Varibar^® barium sulfate trials (Bracco Imaging): three 5-mL thin liquid trials (International Dysphagia Diet Standardization Initiative “IDDSI” Level 0) from a 30-ml medicine cup (40% w/v ratio); comfortable, self-selected cup sip of 90-ml of thin liquid (IDDSI Level 0) from a cup (Solo Clear Graduated Medicine Cups); consecutive cup sips to finish the amount of thin liquid remaining following the comfortable cup sip (IDDSI Level 0); three 5-ml moderately thick trials from a tablespoon (IDDSI Level 3); two 15 ml trials of extremely thick (IDDSI Level 4) pudding from a tablespoon; ¼ graham cracker (Honey Maid, Mondelez Global LLC) with pudding (IDDSI Level 7); and a 13 mm E-Z-Disk^™ barium tablet. All trials except the sequential swallow challenge trial were cued with the instruction to hold the bolus and then swallow. Patients self-administered each bolus trial; however, if the patient was unable to self-administer boluses, the research clinician assisted. To maintain the integrity of subsequent analyses, the research clinicians would implement a liquid wash (water) in between bolus trials if residue was visualized. It is important to note that the goal of this liquid wash was to prevent residue being visible to the rating team; however, this wash did not necessarily ensure a “clean system” as it is still possible that patients with visualized residue on bolus trials may have also had residue from secretions and/or the liquid wash itself.^35–37 Patient safety was maintained throughout the VF by using a bailout criterion. In accordance with these criteria, following the second instance of aspiration on the same bolus type, the third trial would not be administered, and the participant would be given the next thickest consistency. In the event of a third aspiration occurrence or if residue in the valleculae or pyriform sinuses accumulated greater than 75% and was unable to be cleared, the VF was terminated.

2.2.3 |. Analyses and outcomes of interest

Post hoc analyses of all videofluoroscopic data were performed by two independent raters using ImageJ software (National Institutes of Health). For intra-rater reliability purposes, 10% of both safety and efficiency ratings were randomly selected and re-rated by the first author. Individual bolus trial clips were blinded and presented in a randomized fashion (one bolus per clip) to the independent raters. For PAS ratings, inter-rater agreement required 100% consensus between duplicate raters and in any discrepant cases, a consensus meeting was held. For timing outcomes, we followed prior methodology of Smaoui, Peledeau-Pigeon and Steele³⁸ that used a 3-frame discrepancy allowance (i.e., 166.67 miliseconds). Finally, for pixel-based outcomes, a threshold of 2% for valleculae ratings and a threshold of 2.4% for pyriform sinus and extra pharyngeal residue were selected and ratings beyond this boundary were sent to consensus.

The raters were trained by completing a standardized training program specific to our laboratory which included completing reliability training to obtain rating competency (≥90% accuracy) on a benchmark set of previously double-rated clips of ALS patient swallows. As part of the rating protocol, raters were instructed to visually inspect each clip for the presence of residue before rating and, if noted, to only rate new airway invasion and residue events.

2.2.4 |. Maximum anterior isometric lingual pressure

Given that our independent variable of interest was an individual’s maximum anterior isometric lingual pressure, the highest MAIP (i.e., maximum tongue pressure) value recorded across lingual testing trials was used for subsequent data analysis.

2.2.5 |. Swallowing safety

The validated Penetration-Aspiration Scale (PAS)³⁹ was used to determine swallowing safety. The PAS is an eight-point scale that indicates the depth of airway invasion and the patient’s response. PAS scores range from 1 (no airway invasion) to 8 (silent aspiration). Two independent blinded raters completed PAS ratings on every elicited swallow for each bolus trial. Swallowing safety classifications were derived and were defined as Safe (PAS: ≤2), Penetrators (PAS: 3–5), and Aspirators (PAS: 6–8). In accordance with previously described methods,^27,34,40,41 the worst PAS score across all trials (excluding consecutive cup sips) and a safety classification were used in statistical analysis. To perform a binary logistic regression analysis, we had to derive two safety classifications as the outcome for this analysis is categorical (dichotomous). Safety classifications were defined as Safe (PAS 1, 2, 4) and Unsafe (PAS 3, 5, 6, 7, 8) based on categorical groupings recommended by Steele⁴² (these classifications represent “PAS A” vs. “PAS B, C, and D”).

2.2.6 |. Swallowing efficiency

Vallecular, pyriform sinus, and extra pharyngeal space residue was analyzed using the residue subcomponent of ASPEKT method. Residue methodology is described in full detail in our prior publication³⁴ and in the original ASPEKT publication by Steele and Colleagues.⁴³ Residue was analyzed on the first trial of the 5-ml thin liquid, 5-ml moderately thick liquid, and comfortable cup sip of thin liquid for a total of three bolus trials analyzed. Binary efficiency classifications were determined and defined as Efficient: worst total pharyngeal residue <3% and Inefficient: total pharyngeal residue ≥3%. A 3% threshold was used given that recent data suggest that in healthy persons, this represents the established upper limit (two standard deviations) of residue.²⁸

2.2.7 |. Swallowing timing

Several timing metrics pertaining to swallow reaction and laryngeal vestibule closure (LVC) were selected for analysis to better characterize relationships between swallowing safety and MAIP. These metrics were defined and rated in accordance with the Analysis of Swallowing Physiology: Events, Kinematics and Timing (ASPEKT) method decribed in full detail by Steele and colleagues⁴³ and prior work by Waito, Mancopes, and Gandhi.^44–46 The specific timing metrics for the current study included (1) LVC Duration (amount of time in milliseconds (msec) between LVC onset and laryngeal vestibule opening); (2) Time-to-LVC (hyoid burst to onset of LVC in msec); and (3) Swallow Reaction Time (interval between bolus passing ramus of mandible and onset of LVC in msec). These timing metrics were analyzed on the first swallow of the 5-ml thin liquid, cup sip liquid, and 5-ml moderately thick bolus trials for a total of nine timing ratings per participant.

2.3 |. Statistical analysis

SPSS Statistics for Windows (IBM Corp.) and GraphPad Prism Version 9.4.1 for Windows (GraphPad Software) were used for statistical analyses with an α set at 0.05. Descriptive statistics were calculated to characterize cohort demographics and lingual pressure, swallowing safety and swallowing efficiency profiles. Binary logistic regression analyses were used to evaluate relationships between MAIP with swallowing safety and swallowing efficiency. Spearman’s rho correlation analyses (r_s) were used to assess associations between MAIP with swallowing efficiency and timing metrics. Cohen’s kappa was used to determine inter-rater and intra-rater reliability for safety ratings. Two-way mixed intraclass correlation coefficients (ICCs) were generated to examine intra- and inter-rater reliability for efficiency ratings. Level of agreement for intra- and inter-rater ratings was determined in accordance with published guidelines.^47,48

3 |. RESULTS

Demographic participant outcomes are summarized in Table 1. Frequency distributions for PAS scores, residue percentages, and timing metrics are provided in the supplemental material (Figure S1 and Tables S1–S4). Nine participants (9.3%) reached bailout criterion with two of these participants (2.2%) having their VF terminated due to three instances of aspiration and excessive accumulation of residue, respectively. The remaining seven participants (77.8%) moved to the next thickest consistency. Worst PAS scores and worst total residue percentages were derived for each participant. Timing analysis was undertaken for a subset of 82 participants (84.5%). Among this subset, LVC duration analysis was not completed on 3 (3.7%), 8 (9.8%), and 6 (7.3%) participants; time-to-LVC was not completed on 3 (3.7%), 8 (9.8%), and 5 (6.1%) participants; and swallow reaction time was not completed on 2 (2.4%), 9 (10.9%), and 6 (7.3%) participants across the 5 ml thin liquid, cup sip thin liquid, and 5 ml moderately thick boluses, respectively. These timing ratings were unable to be completed due to issues encountered while completing analysis such as truncated videos, obstructed view of structures, and bolus not administered. No participants were excluded from analysis due to missing data; rather, analyses were performed on available data for each participant.

Table 1.

Demographic characteristics of study cohort (N=97).

Demographic Variable:	Mean (SD)	Minimum	Maximum
Age (years)	63 (10.8)	28	85
Disease Duration (months † )	24.9 (20.6)	5	109
ALSFRS-R† Total Score	35.1 (7.6)	16	48
ALSFRS-R† Bulbar Score	9.0 (2.4)	3	12
Sex	48% Male 52% Female
Onset Type	47.9% Bulbar 46.8% Spinal 5.3% Mixed
Race	88.6% White or Caucasian 5.2% Black or African American 1.0% Asian 2.1% Native Hawaiian or Other Pacific Islander 3.1% Unknown

^†ALSFRS-R – Amyotrophic Lateral Scale Functional Rating Scale-Revised.⁷³

Disease duration listed from time since symptom onset.

3.1 |. VF rater reliability

For PAS VF ratings, “substantial” agreement was noted for intra- (κ = 0.75; 95% CI: 0.45–1.04) and inter-rater reliability (κ = 0.72; 95% CI: 0.60–0.84), p < 0.05. For VF efficiency ratings, “good” to “excellent” reliability was noted for intra- (ICC: 0.89; 95% CI: 0.80–0.94) and inter- (ICC: 0.90; 95% CI: 0.86–0.93) rater reliability, p < 0.05.

3.2 |. MAIP in ALS

Table 2 summarizes participant lingual pressure profiles across the entire cohort and within established age divisions.⁴⁹ Mean MAIP for this cohort was 36.3 kPa (SD: 18.7).

Table 2.

Summary of maximum anterior isometric lingual pressure (MAIP) profiles stratified by age groups and compared to expected normative data.⁴⁹

LPC (kPa)	Mean	Expected	% Predicted
Entire Cohort	36.3 (18.7)
Young (n=28)	40.4 (20.9)	62.4kPa	64.7%
Old (n=69)	34.6 (17.7)	57.4kPa	60.3%

Norms: Young = <60 years, Old = ≥60 years.

3.3 |. MAIP and swallowing safety

Mean MAIP and standard deviation across swallowing safety groups are depicted in Figure 1. The binary logistic regression model was significant (G = 18.5, df = 1, p < 0.0001). The model explained 17.3% of the variance (Cox-Snell R²)⁵⁰ in swallowing safety and correctly classified 73.9% of cases. The odds of safe swallowing was increased with increasing MAIP (OR: 1.06, 95% CI: 1.03–1.09; Figure 2).

FIGURE 1.

(A) Bar graphs depicting mean (standard error) lingual pressure physiologic capacity (MAIP) for each swallowing safety group and including representative images of (B) safe swallowing (Penetration Aspiration Scale³⁹ (PAS):1–2), (C) penetration (red arrow; PAS 3–5), and (D) aspiration (blue arrow; PAS 6–8).

FIGURE 2.

(A) Bar chart depicting means and standard error of lingual pressure physiologic capacity (MAIP; kPa) across swallowing efficiency binary classifications: (B) efficient: worst total residue <3%C2–C4²; (C) inefficient: worst total residue ≥3%C2–C4.²⁴³

3.4 |. MAIP and swallowing efficiency

Mean MAIP and standard deviation across swallowing efficiency groups are summarized in Figure 3. Significant correlations were observed between MAIP and residue in the valleculae (r_s = −0.26), pyriform sinuses (r_s = −0.26), and extra pharyngeal spaces (r_s = −0.34), p < 0.05. The binary logistic regression model investigating the relationship between MAIP and swallowing efficiency was significant (G = 7.4, df = 1, p = 0.006). The model correctly classified 68.9% of cases and explained 7.4% (Cox-Snell R²) of variance in swallowing efficiency. Increased MAIP was associated with an increased odds of efficient swallowing (OR: 1.04; 95% CI: 1.01–1.06, Figure 4).

FIGURE 3.

Results of a binary logistic regression demonstrating the odds safe or unsafe swallowing as a function of maximum anterior isometric lingual pressure (MAIP; kilopascals (kPa)). Binary safety groups included Safe: Penetration-Aspiration Scale (PAS)³⁸ scores 1, 2, and 4; Unsafe: PAS scores 3, 5, 6, 7, and 8 and are depicted on the y-axis by 1 and 0, respectively. A one-unit (+1 kPa) increase in MAIP was associated with a 1.06 increased odds of being a safe swallower (95% CI: 1.03–1.09).

FIGURE 4.

Graphical results of binary logistic regression analysis performed to examine the effect of maximum anterior isometric lingual pressure (MAIP; kPa) on efficiency status. Binary efficiency groups are depicted on the y-axis as 0 (Inefficient: worst total C2–C4² residue ≥3%) and 1 (Efficient: worst total C2–C4² residue ≤3%).⁴² A MAIP decrease of 1 kPa was associated with a 1.04 increased odds of being an efficient swallower (95% CI: 1.01–1.06).

3.5 |. MAIP and swallowing timing

Table 3 summarizes results for the correlation analyses performed between swallowing timing metrics and MAIP. Significant, negative correlations were noted between time-to-LVC and MAIP on the 5 ml thin liquid (r_s = −0.35, p = 0.002), cup sip thin liquid (r_s = −0.26, p = 0.02), and 5 ml moderately thick (r_s = −0.28, p = 0.01) bolus types. A negative association was also observed between LVC duration on the cup sip thin liquid and MAIP (r_s = −0.34, p = 0.003). No other significant relationships between swallow timing and MAIP were noted.

Table 3.

Summary of Spearman’s rho correlation coefficients (r_s) examining relationships between maximum anterior isometric lingual pressure and temporal metrics of swallowing in this cohort of 97 individuals with amyotrophic lateral sclerosis.

Swallowing Timing Metric:	5-mL Thin:		Cup Sip Thin:		5-mL Moderately Thick:
	rs	P-Value	rs	P-Value	rs	P-Value
LVC Duration (msec)	0.02	0.84	−0.34	0.003*	−0.02	0.85
Time-to-LVC (msec)	−0.35	0.002*	−0.26	0.02*	−0.28	0.01*
Swallow Reaction Time (msec)	−0.04	0.76	0.05	0.71	0.01	0.93

Asterisks (*) denote significance at the p < 0.05 level.

4 |. DISCUSSION

We examined MAIP profiles in 97 individuals with ALS and sought to examine potential relationships between MAIP, temporal aspects of deglutition as well as swallowing safety and efficiency. As hypothesized, MAIP was reduced in this group of individuals with ALS compared to expected normative values. Lower MAIP was associated with an increased odds of unsafe swallowing and inefficient swallowing. Individuals with reduced MAIP also demonstrated longer time-to-LVC time (thin boluses) and prolonged LVC duration (cup sip thin). Swallowing reaction time was not associated with MAIP. Collectively, these data highlight that reduced lingual pressure in ALS yields important associations with functional aspects of swallowing safety, efficiency, and timing.

Average MAIP in this group of individuals with ALS was 38 kPa, representing 63% of the expected force generating capacity in healthy adults⁴⁹ and falling below the reported minimal lingual pressure threshold of 40 kPa for aspiration risk in healthy adults.^20,23 Our findings are consistent with reports from other neurodegenerative populations, including Parkinson’s Disease¹⁷ and oculopharyngeal muscular dystrophy,⁵¹ where similar reductions in lingual strength are noted. The hypoglossal motor neuron degeneration and morphological lingual changes occurring in ALS^7–9,52 provide a structural basis and explanation for reduced lingual pressure generation.

Reduced lingual pressure (MAIP) was associated with airway invasion, in a dose-dependent fashion with greater degrees of airway invasion observed with greater depletion in MAIP. This work is in agreement with our recent finding that maximum lingual pressures <43 kPa are associated with unsafe swallowing in ALS.⁵³ Indeed, our current study found that the average MAIP in the safe swallowing group was 43.2 kPa while penetrators and aspirators demonstrated MAIPs that were on average 12.5 kPa and 17.5 kPa below ALS participants with safe swallowing, respectively. Further, across the thin bolus trials, a longer time-to-LVC was associated with lower MAIP and for the thin cup sip trial, reduced MAIP was associated with a prolonged LVC duration. An increased time to close the laryngeal vestibule during swallowing may reduce swallowing safety by allowing time for material to enter the airway. While we did not hypothesize we would find associations between MAIP and LVC duration, this finding may suggest that reduced tongue strength in ALS is reflective/associated with declines in components of LVC such as hyolaryngeal movement^54–56 and pharyngeal constriction⁵⁷ which are noted to occur during disease progression. Impairments to these structures likely underlie aberrant mechanisms of LVC that have been previously reported.^58–60 Several kinematic and temporal swallowing metrics such as swallow and laryngeal reaction are contingent upon timely movement of the hyoid bone.⁶¹ While the tongue is a muscular hydrostat generally free from traditional skeletal support,⁵ the root of the tongue is affixed to the hyoid bone. Previously, a rostrocaudal progression of impairment in ALS has also been suggested wherein structures appearing earlier in the oropharynx such as the lips are affected earlier in the disease course than later appearing bulbar structures such as the pharyngeal muscles.⁶² The noted changes in MAIP may be indicative of physiologic decline of the floor of mouth muscles situated directly below the tongue (mylohyoid, geniohyoid, digastric, stylohyoid) which require adequate strength to contract and move the hyoid in a timely manner during swallowing.⁶³ In addition, during the pharyngeal phase of swallowing, tongue base retraction, in concert with epiglottic inversion and pharyngeal constriction, facilitate closure of the laryngeal vestibule, which is crucial for airway safety.⁶ We did not measure posterior lingual pressures in the current study. As such, future work that also incorporates simultaneous VF and high-resolution manometry during swallowing could be considered to measure the relationships between posterior tongue pressures, base of tongue retraction, and airway protection in ALS to better understand if reduced anterior lingual pressures are associated with changes to posterior lingual pressures and the ability to retract the tongue.

Similar to recent reports,^34,44 a very high proportion (approximately three quarters) of ALS patients demonstrated inefficient swallowing. Participants with inefficient swallowing demonstrated MAIPs significantly lower than those without efficiency impairment. Both groups were also noted to fall below the functional lingual pressure threshold of ≤46 kPa associated with inefficient swallowing that our group recently identified.⁵³ However, the difference (1.2 kPa) between the 46 kPa threshold and mean MAIP in the efficient group in the current study was negligible and may be due to the current study including a larger sample of patients. Further, reduced lingual strength was associated with increased residue across all anatomical sites, with residue in the extra pharyngeal spaces demonstrating the strongest association with MAIP. Given the paucity of investigations that have indexed efficiency outcomes,⁶⁴ these preliminary data highlight the need for a greater focus on this important aspect of swallowing that has historically been under-studied as compared to safety.

Decreased lingual pressure generation has previously been associated with inefficient bolus transport and vallecular residue in the elderly⁶⁵ and acquired brain injury patient⁶⁶ populations. It is likely that individuals with reduced lingual pressure also have a diminished ability to generate adequate force behind the bolus during swallowing to facilitate efficient bolus propulsion and clearance from the oral cavity into the pharynx. The current findings of prolonged LVC duration and time-to-LVC noted in participants with reduced MAIP combined with previous reports of impaired pharyngeal constriction⁵⁷ and LVC timing and kinematics in ALS⁵⁸ collectively suggest a decreased ability to clear pharyngeal residue in ALS, thus increasing risk of subsequent airway invasion into an open laryngeal vestibule. More research directly evaluating relationships and underlying mechanisms of dysphagia in ALS is warranted.

Although swallowing is a submaximal task, utilizing only ~38% (young) or 54% (older) of maximum lingual pressure,⁶⁷ reduced MAIP impacts lingual physiologic reserve for swallowing and the relative percentage effort required for each swallow attempt. This is due to the way physiologic reserves are traditionally conceptualized with the upper physiologic boundary representing the maximum capacity of a system or structure while the lower boundary represents the amount of physiologic capacity used for basal function.^68,69 Thus, a loss in MAIP (upper bound) will affect the overall amount of reserve available. These factors have important clinical implications for swallowing efficiency. As previously highlighted,¹³ the loss of lingual “pressure reserve” (i.e., lingual physiologic reserve) may impact the ability to modulate lingual pressures to tolerate a wide variety of textures and consistencies, thereby affecting diet recommendations.¹³ Some individuals with ALS are also known to function in a hypermetabolic state;⁷⁰ thus, conservation of energy within this patient population is of utmost importance. However, reduced lingual strength is associated with prolonged mealtimes¹⁹ and increased swallowing frequency during swallowing in ALS.⁷¹ The reduced MAIP in ALS, therefore, creates a concerning and paradoxical situation wherein patients have a metabolic requirement for increased nutritional intake— in the context of swallowing that is progressively more inefficient and unsafe. The impairments in swallowing contribute to an imbalance in energy and nutritional homeostasis that places an ALS patient at a significant physiological disadvantage. Together, this places patients with ALS into a vicious cycle in which physiologic reserves are depleted, further contributing to disease progression and decline.

The current study combined with our previous investigation⁵³ suggests that MAIP could represent a useful and pragmatic clinical marker to be probed and monitored over time in ALS clinical settings. Indeed, a recent bulbar expert consensus practice recommendation statement by the Northeast ALS Consortium recommended evaluation and tracking of lingual pressure testing as one domain of the clinical evaluation.⁷² Therefore, clinicians may consider using lingual pressure testing in conjunction with other subjective and objective data obtained from patient visits to guide decision-making and requests for instrumental assessment referrals.

There are several important study limitations. First, the methodology used in this investigation focused on the upper physiological boundary of lingual physiologic reserve, maximum lingual pressure. However, lingual physiologic reserve encompasses the facets of both maximum tongue strength and saliva swallowing pressures. Although previous research in ALS has measured both maximum and saliva swallowing pressures in ALS,¹³ no functional swallowing outcomes were associated with this study. Therefore, it remains unclear how submaximal tongue generating forces and total lingual physiologic reserve are related to swallowing safety, efficiency, and timing in this patient population. Future work should focus on identifying the relationships between the multifaceted aspects of lingual physiologic reserve capacity and swallowing function in ALS. Second, these data were reported from a single patient visit (i.e., cross-sectional study design). However, ALS is a progressive neurodegenerative disease; therefore, a longitudinal study is necessary to better understand the progression of lingual and swallowing impairments in ALS. In addition, while we did collect demographic information pertaining to race, other outcomes related to diversity, inclusion, and equity such as ethnicity, socioeconomic status, etc. were not included in the current study. More research is needed to understand how these underlying contextual factors and sociodemographic determinants of health may influence the observed relationships.

These data demonstrate that reduced MAIP was associated with both unsafe and inefficient swallowing in individuals with ALS. Given these findings, inclusion of MAIP testing during the bulbar mechanism examination in individuals with ALS is recommended. Our ongoing natural history study will enable observation and a better understanding regarding how an objective MAIP metric tracks with swallowing decline over time.

Key Points.

• Early and extensive physiologic changes to the tongue have been reported in patients with amyotrophic lateral sclerosis (ALS). Therefore, measuring maximum tongue strength may be of clinical interest and relevance to the bulbar assessment in this patient population.

• In this cohort of 97 inidividuals with ALS, decreased maximum anterior isometric lingual pressure (MAIP) was associated with impairments in swallowing safety, efficiency, and timing.

• These results suggest assessing lingual pressure along with other subjective and objective clinical data may yield important insights regarding swallowing function during ALS disease progression.