Cough Correlates of Functional Swallow Outcomes in Atypical Parkinsonism

Abstract

Background

Swallow and cough impairments lead to aspiration and reduced clearance of aspirate material. Both behaviors are impaired in Parkinson's disease, but it is unknown whether a similar relationship of dysfunction exists in forms of atypical Parkinsonism (APD). Elucidating this association in APD may lead to early, comprehensive airway protection treatment.

Objectives

We tested the hypotheses that swallow deficits in APD are associated with impaired cough and that airway protective dysfunction is associated with longer disease duration.

Methods

Swallowing difficulty was described by 11 participants with APD. Penetration‐Aspiration Scale (PAS) and DIGEST scores for thin liquid trials were extracted from medical records of videofluoroscopic swallow study reports. Voluntary and capsaicin induced‐reflex cough measures of flow, volume, and timing were analyzed.

Results

While most participants did not have post‐swallow residue, ~80% received abnormal PAS scores and reported swallowing difficulty. Those with abnormal PAS scores had lower voluntary cough expired volume (P = 0.037; mean rank difference = 5.0); lower reflex inspiratory flow rate (P = 0.034; mean rank difference = 5.5); and longer reflex expiratory flow rise time (P = 0.034; mean rank difference = 5.5). Higher PAS scores and reduced reflex cough volume acceleration were significantly correlated (r = −0.63; P = 0.04) and longer disease duration predicted larger voluntary cough expired volume (R² = 0.72) and longer flow rise times (R² = 0.47).

Conclusions

As swallow safety worsens, so might the ability to clear the airways with effective cough in in APD; particularly with longer disease duration. Assessing cough in conjunction with swallowing is important for informing airway protection treatment plans in APD.

Swallow and cough disorders are often concomitant symptoms of neurological disease, ¹ , ² , ³ , ⁴ , ⁵ , ⁶ since both functions share central neural substrates and operate within the aerodigestive tract. ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² Disordered cough (dystussia) can compound the effects of disordered swallowing (dysphagia), increasing the risk for patients to develop fatal respiratory complications, ¹³ , ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ experience decreased quality of life, ¹⁸ , ¹⁹ and undergo a decline in nutrition and hydration status. ¹⁸ Decline in function and complications of disease also increase the frequency of hospitalizations, healthcare costs, burden of care, and mortality. ²⁰ , ²¹ Thus, it is critical to evaluate swallow and cough to promptly initiate rehabilitative strategies that promote health.

Airway protective dysfunction ⁷ is particularly evident in Parkinson's disease (PD). The slow propagation of Lewy body inclusions across neural substrates, and the reduction of dopaminergic modulation, impair sensorimotor functions ²² , ²³ —important components of breathing‐swallowing coordination, ²⁴ , ²⁵ , ²⁶ , ²⁷ bolus transport, ²⁸ , ²⁹ , ³⁰ laryngeal vestibule closure, ³¹ , ³² , ³³ and pharyngeal muscle activation during swallowing. ³⁴ , ³⁵ These swallowing‐related deficits can result in aspiration, ¹ , ² , ³ , ⁴ , ⁵ , ⁶ , ³⁶ , ³⁷ which should stimulate a cough; however, cough may be absent or ineffective due to disease impacts on cough control centers.

Cough impairments in PD stem from poor sensorimotor integration, chest wall rigidity, uncoordinated motor control, and/or respiratory muscle weakness. ³⁸ , ³⁹ Based on airflow data, specific cough deficits include atypical compression phase durations; reduced peak expiratory flow rates; prolonged expiratory flow rise times; and low cough expired volumes. ⁴⁰ , ⁴¹ , ⁴² Varying degrees of dysfunction hinder the ability to effectively eject aspirate material from the airways, such that less force compresses the chest wall to quickly drive air from the lungs; an important physiological component of cough. ⁴³ Importantly, persons with PD who have impaired cough airflows and disorganized patterns of sequential cough expired volume are more likely to have swallowing deficits with aspiration. ⁴⁰ , ⁴¹ , ⁴² , ⁴⁴ , ⁴⁵

There is evidence to suggest that rare, rapidly degenerating, atypical Parkinsonism (APD) (eg, Progressive Supranuclear Palsy and Multiple System Atrophy) results in similar swallowing deficits as PD, leading to airway compromise that worsens with disease progression and increases the risk for fatal respiratory infections. ⁴⁶ , ⁴⁷ , ⁴⁸ , ⁴⁹ , ⁵⁰ , ⁵¹ , ⁵² , ⁵³ , ⁵⁴ , ⁵⁵ , ⁵⁶ , ⁵⁷ Few studies have investigated dystussia in APD, despite evidence that disease pathologies impact cortical and brainstem centers involved in the control of breathing and cough. ⁵³ , ⁵⁴ , ⁵⁵ , ⁵⁶ , ⁵⁷ Recent findings suggest that voluntary and induced‐reflex cough are severely impaired in APD, specifically in Progressive Supranuclear Palsy ⁵⁸ , ⁵⁹ ; however, it is unknown whether these cough metrics are related to the presence of penetration or aspiration.

Because swallow and cough deficits greatly impact health, elucidating potential relationships could optimize diagnostic methods for earlier identification of deficits and earlier implementation of salient treatment plans—important goals given the short expected lifespan after diagnosis in APD. ¹⁴ , ¹⁵ , ⁶⁰ This study aimed to test the hypotheses that abnormal penetration and/or aspiration of liquid during swallowing is associated with worse cough airflows in people with APD. Because disease duration is correlated with greater cough and swallow dysfunction in PD¹, we also hypothesized that longer disease duration is associated with worse airway protection outcomes in APD.

Methods

Participants

Participants with a clinical diagnosis of possible or probable Progressive Supranuclear Palsy (PSP) or Multiple System Atrophy (MSA) were recruited from the academic health institution where they received clinical care. They volunteered to participate in a larger study examining sensorimotor cough characteristics in APD (IRB202101196). Diagnoses were based on Hoglinger et al ⁵⁵ and Gilman et al ⁶¹ and provided by neurologists who were fellowship trained in movement disorders.

Participants were included in the analyses if they had a videofluoroscopic swallow study within one year of cough evaluation. We considered a 6‐month time‐frame; however this would have resulted in a smaller sample. We chose to expand criterion to one year to gain a broader understanding of airway protection deficits in a larger cohort of rare disease. Exclusion criteria included (1) greater than five years since diagnosis, (2) inability to ambulate at least 10 feet (with/without assistance), (3) other neurological diagnoses, (4) history of head, neck, or lung cancer, (5) current respiratory infection, (6) chronic respiratory disease, (7) cigarette smoking within five years, (8) allergy to capsaicin, (9) Montreal Cognitive Assessment score less than 20, or (10) severe/profound risk for depression based on the Beck Depression Inventory II. ⁶²

Participants provided verbal and written informed consent. Study procedures were approved by the University's Institutional Review Board.

Procedures

Medical History Information

Participants completed the Montreal Cognitive Assessment, Beck Depression Inventory‐II, and a questionnaire regarding medical history and descriptions of swallowing difficulty (eg, “do you experience difficulty swallowing? If so, please describe.”). Disease duration since symptom onset and since medical diagnosis, disease severity rating (PSP Rating Scale ⁶³ ; Unified MSA Rating Scale ⁶⁴ ; Unified PD Rating Scale ⁶⁵ ), biological sex, and age were obtained from electronic medical records.

Videofluoroscopic Swallow Study

The research team collected swallow outcomes from medical records of clinical videofluoroscopic swallow study results that were analyzed by a staff speech‐language pathologist. All imaging studies were completed in the same clinic using the OEC9900 Elite, GE Healthcare, E003800 C‐arm. Videos were recorded at 30 frames per second, in the lateral viewing plane. The standard clinical protocol included 2–3 single sips (10 mL) of thin liquid barium (Varibar, UltraThin Barium), a 90 mL thin liquid sequential drinking task, two swallows of pudding, and a solid swallow task consisting of a cracker coated with barium pudding. For the current study, only thin liquid barium swallows (International Dysphagia Diet Standardization Initiative Level 0 ⁶⁶ ) were included in the analyses. Instructions for the 10 mL swallows were to “hold the bolus in your mouth until you are ready to swallow,” and for the 90 mL task, “drink all the liquid in the cup without stopping.” The primary outcome was the highest Penetration‐Aspiration Scale (PAS) ⁶⁷ score recorded in the clinical report during the 10 and 90 mL drinking tasks. The Dynamic Imaging Grade of Swallowing Toxicity (DIGEST) ⁶⁸ efficiency score was the secondary outcome measure.

Voluntary Cough Testing

Cough spirometry was collected using a pneumotachograph (MLT1000L, ADInstruments) with differential pressure transducer (FE141, ADInstruments), connected to a disposable facemask (Ambu, 000252055). Spirometry volume was calibrated in LabChart8 software (ADInstruments) using a 3 L syringe. The facemask was placed securely over the participant's nose and mouth. Participants acclimated to the equipment with one minute of rest breathing before they were asked to cough into the facemask “as if something went down the wrong pipe.” They completed this three times, with rest breathing between each trial.

Induced‐Reflex Cough Testing

Participants completed induced‐reflex cough testing using a capsaicin (IND #76866) dose–response method. ⁶⁹ They breathed through the same facemask and spirometry equipment as during the voluntary cough task set up. However, during reflex cough testing, a nebulizer (DeVilbiss Healthcare, LLC, 6462017‐06‐21) containing a saline control, or one of four doses of capsaicin (50, 100, 150, and 200 μM dissolved in 80% physiologic saline and 20% ethanol), was attached in‐line to an inspiratory valve connected to the facemask. A dosimeter (KoKo Dosimeter, Ferraris 2004KD005) was triggered upon inhalation, while a portable air compressor (PulmoMate, DeVilbiss Healthcare, 465OD D7011170) aerosolized one of the five stimulus solutions at a flow rate of 5 L/min for two seconds. The facemask was removed after participants coughed, or after 10 s. The capsaicin and saline control doses were presented in randomized order, across three blocks, and participants were instructed to “cough if you need to.”

Peak expiratory flow rate was the primary outcome for voluntary and reflex cough. Secondary outcomes included: peak inspiratory flow rate, inspiratory volume, compression phase duration, peak expiratory flow rise time, cough volume acceleration, and cough expired volume. One rater analyzed all voluntary and induced‐reflex cough measures. The same rater re‐analyzed 20% of the data to assess intra‐rater reliability. A second rater analyzed 20% of the data to determine inter‐rater reliability.

Statistical Analyses

SPSS v. 27 and Prism 9 were used for statistical analyses and figures. Descriptive statistics were used to summarize participant demographics, as well as primary and secondary outcome measures. Due to non‐normal distribution and small sample size, nonparametric tests were used. To address the first aim, Kruskal Wallis tests compared voluntary and reflex cough airflows between two PAS groups (participants with PAS scores ≥3 versus scores <3) for each thin liquid bolus trial (10 and 90 mL thin liquid). Upon noted differences, we sought to elucidate which cough outcomes were associated with higher PAS scores using bivariate correlational tests. To address the second aim, linear regression models were used to determine whether longer disease duration was associated with worse swallow and cough outcomes. Intraclass correlational analyses assessed inter‐ and intra‐rater reliability using two‐way random effects, absolute agreement on single measures. A priori alpha level was set for P < 0.05.

Results

Participant Characteristics

Eleven participants (3 female) completed a videofluoroscopic swallow study within one year of voluntary and reflex cough assessments; 8 within 6 months and 3 within one year. Ten participants were diagnosed with PSP; one with MSA. Table 1 provides a summary of participant demographics.

TABLE 1.

Summary of descriptive data for disease diagnoses (subtypes of atypical forms of parkinsonism) and other demographics

Forms of Atypical Parkinsonism
	N (%)	Disease rating scale
PSP‐RS	9 (82%)	PSPRS (n = 5); UPDRS (n = 4)
PSP‐PGF	1 (0.09%)	PSPRS
MSA‐P	1 (0.09%)	UMSARS
Use of levodopa	6 (55%)

Participant characteristics
	Mean	SD	Range
Age (years)	70	6.21	60–81
Disease duration since symptom onset (years)	4.66	1.82	2–8
Disease duration since diagnosis (years)	1.84	1.06	0.5–4
Montreal Cognitive Assessment Score	25.09	2.63	20–28
Beck Depression Inventory Score	10.64	7.31	1–28

Abbreviations: PSP, Progressive Supranuclear Palsy; RS, Richardson Syndrome: PGF, Progressive Gait Freezing; MSA‐P, Multiple System Atrophy‐Parkinsonian type; PSPRS, Progressive Supranuclear Palsy Rating Scale; UPDRS, United Parkinson Disease Rating Scale; UMSARS, United Multiple System Atrophy Rating Scale.

Functional Swallow Measures & Participant Descriptions of Swallow Difficulty

Out of 11 participants, nine reported swallowing difficulty described as “foods not going down,” “food getting stuck,” and “choking on liquids.” Figure 1 and Table 2 display descriptive summaries regarding PAS and DIGEST efficiency scores for thin liquid boluses, as well as reports of perceived swallowing difficulty.

Figure 1.

Percentage of participants with no airway compromise (PAS scores 1–2), penetration (scores 3–5), and aspiration (scores 6–8) during 90 and 10 mL thin liquid barium trials during videofluoroscopy.

TABLE 2.

Descriptive reports from participants with atypical Parkinsonism, after they were asked to describe their perceived, overall swallowing difficulty

Participant number	DIGEST Efficiency Score	Participant self‐report of swallowing difficulty
01	0	“Some foods don't go down as they should.”
02	0	“Occasional coughing with solid, dry foods.”
03	0	“Foods stick but sips of water help.”
04*	4	“Sticking with thick chunky consistencies”
05	1	“Thick, hard food gets stuck… had to do Heimlich three times.”
06	0	“Crackers and peanuts get stuck sometimes… … “cough and choke with liquids.”
07	0	“Hard to chew food and move it back… drooling.”
08	1	No report of difficulty.
09	Not available	No report of difficulty
10	0	Per spouse, “Coughs more with something like popcorn…[and] when he drinks.”
11	0	“Large pills stick but water helps… coughs on saliva at night.”

Note: Most reports involved difficulty with foods, although three individuals also had difficulty with liquids. The DIGEST was not used as a measure in the swallow study for Participant #9.

^*NOTE: participant with history of severe dysphagia and pneumonia; supplements oral intake with tube feeding.

Abbreviation: DIGEST, Dynamic Imaging Grade of Swallowing Toxicity.

Voluntary and Induced‐Reflex Cough Measures

One participant with probable PSP‐Richardson Syndrome was unable to produce a voluntary cough. He was able to volitionally inspire after given instructions, but could not produce an expulsive cough during expiration. The participant appeared frustrated and reported, “I can't do it.” For those participants who could volitionally cough (n = 10), the first coughs across three trials were averaged. Deficits were noted during the expiratory phase, such that peak expiratory flow rise time was longer (mean = 0.13 s ± 0.08), and peak expiratory flow rate was lower (mean = 1.40 L/s ± 0.68) than historical reports in PD and healthy controls. ⁴¹ , ⁴²

During induced‐reflex cough testing, all participants had at least one cough in response to 200 uM of capsaicin. Measures in the first cough of all three trials at this dose were averaged for analysis and showed deficits in timing, such as prolonged compression phase duration (mean = 0.51 s ± 0.26) and peak expiratory flow rise time (mean = 0.07 s ± 0.05). Table 3 shows summary data for voluntary and induced‐reflex cough.

TABLE 3.

Descriptive data for voluntary and induced‐reflex cough airflows in people with forms of atypical Parkinsonism

	Voluntary cough measures			Induced‐reflex cough airflows
	Mean	SD	Range	Mean	SD	Range
Peak inspiratory flow rate (L/s)	0.83	0.30	0.45–1.44	0.54	0.27	0.27–1.26
Inspiratory volume (L)	0.94	0.38	.25–1.76	0.33	0.10	0.20–0.50
Compression phase duration (s)	0.44	0.25	0.25–1.07	0.51	0.26	0.13–1.02
Peak expiratory flow rise time (s)	0.13	0.08	0.04–0.23	0.07	0.05	0.03–0.16
Peak expiratory flow rate (L/s)	1.40	0.58	0.69–2.56	2.00	0.66	0.41–2.56
Cough volume acceleration (L/s/s)	17.24	12.71	1.07–35.42	42.79	25.76	1.02–82.44
Cough expired volume (L)	0.29	0.23	0.03–0.72	0.24	0.11	0.09–0.42

Intra‐ and Inter‐Rater Reliability

Respectively, voluntary cough intra‐ and inter‐rater reliability coefficients were 0.973 and 0.991 (peak inspiratory flow rate); 0.998 and 0.985 (inspiratory volume); 0.979 and 0.801 (compression phase duration); 0.813 and 0.767 (peak expiratory flow rate); 0.969 and 0.974 (expiratory flow rise time); 0.855 and 0.986 (cough expired volume). Respectively, reflex cough intra‐ and inter‐rater reliability coefficients were 1.00 and 0.996 (peak inspiratory flow rate); 0.990 and 0.980 (inspiratory volume); 0.819 and 0.988 (compression phase duration); 0.939 and 0.944 (peak expiratory flow rate); 1.0 and 0.955 (expiratory flow rise time); 0.861 and 0.848 (cough expired volume).

Associations between Participant & Disease Characteristics, Swallow Outcomes, and Cough

The regression model (Y = 0.1061*X−0.1891) revealed disease duration since symptom onset was strongly related to increased cough expired volume (R² = 0.72, F (1, 8) = 20.72, P = 0.002) and increased peak expiratory flow rise time (R² = 0.47, F (1, 8) = 7.21, P = 0.03) during voluntary cough. Thus, as disease duration increased, so did cough expired volume and flow rise time. Disease duration was not associated with reflex cough outcomes (Table 4).

TABLE 4.

Regression model results with disease duration since symptom onset as the independent predictor

	Voluntary cough measures				Induced‐Reflex cough airflows
	F statistic	R2	P value	Slope	F statistic	R2	P value	Slope
Peak inspiratory flow rate (L/s)	1.713	0.20	0.23	−0.07	0.77	0.07	0.40	−0.04
Inspiratory volume (L)	0.20	0.02	0.67	−0.03	2.59	0.22	0.14	−0.02
Compression phase duration (s)	0.06	0.01	0.82	0.01	1.96	0.18	0.20	−0.06
*Peak expiratory flow rise time (s)	7.21	0.47	0.03	0.03	0.12	0.002	0.90	−0.001
Peak expiratory flow rate (L/s)	0.01	0.001	0.94	−0.01	2.22	0.20	0.17	−0.16
Cough volume acceleration (L/s/s)	3.13	0.28	0.11	−3.63	0.34	0.04	0.57	−2.70
**Cough expired volume (L)	20.72	0.72	<0.01	0.11	0.001	0.0002	0.98	0.001

Note: Significant associations were noted with peak expiratory flow rise time (*P < 0.05) and cough expired volume (**P < 0.01) for voluntary cough. No significant associations were with disease duration and induced‐reflex cough outcomes.

PAS group had no effect on voluntary and induced‐reflex cough measures in 10 mL thin liquid boluses. In the 90 mL trials, those with PAS ≥3 had significantly greater voluntary cough expired volume (H (1)=4.36; P = 0.037; mean rank difference = 5; Fig. 2A), as well as lower peak inspiratory flow rate (H (1)=4.50; P = 0.034; mean rank difference = 5.5; Fig. 2B) and longer peak expiratory flow rise time (H (1)=4.50; P = 0.034; mean rank difference = 5.5; Fig. 2C) for reflex cough. Those with PAS ≥3 had reduced reflex cough volume acceleration (mean rank difference = 4.89; Fig. 2D and Table 5), although the difference was not statistically significant.

Figure 2.

Differences between PAS groups for voluntary and reflex cough at 90 mL thin liquid bolus. PAS ≥3 had significantly (A) greater voluntary cough expired volume, (B) lower reflex inspiratory flow rate, and (C) longer reflex expiratory flow rise time. Reflex cough volume acceleration was lower in the PAS ≥3 group; however this difference was nonsignificant.

TABLE 5.

Descriptive statistics and Kruskal Wallis H test results comparing voluntary cough (VC) and reflex cough (RC) airflows between PAS score groups (≥3 vs. < 3) at 90 mL of thin liquid

	PAS ≥3	PAS <3
	Mean (SD)	Mean ± SD	H statistic	P value
PIFR
VC:	0.833 (0.330)	0.792 (0.247)	0.068	0.794
RC:	0.442 (0.114)	0.977 (0.401)	4.50	0.034
IV
VC:	0.925 (0.425)	0.992 (0.172)	0.273	0.602
RC:	0.297 (0.087)	0.459 (0.051)	3.58	0.059
CPD
VC:	0.425 (0.279)	0.468 (0.073)	1.10	0.295
RC:	0.572 (0.251)	0.246 (0.171)	2.72	0.099
PEFR
VC:	1.47 (0.592)	1.09 (0.570)	1.39	0.239
RC:	1.91 (0.700)	2.43 (0.066)	0.502	0.478
PEFRT
VC:	0.285 (0.442)	0.065 (0.031)	2.09	0.148
RC:	0.143 (0.216)	0.033 (0.003)	4.50	0.034
CVA
VC:	15.89 (12.28)	22.64 (18.07)	1.09	0.296
RC:	35.59 (22.28)	75.16 (10.30)	3.56	0.059
CEV
VC:	0.351 (0.220)	0.050 (0.033)	4.36	0.037
RC:	0.221 (0.117)	0.283 (0.011)	0.889	0.346

Note: Bolded values denote VC and RC airflows that were significantly different between the groups. PIFR, peak inspiratory flow rate (L/s); IV, inspiratory volume (L); CPD, compression phase duration (s); PEFR, peak expiratory flow rate (L/s); PEFRT, peak expiratory flow rise time (s); CVA, cough volume acceleration (L/s/s); CEV, cough expired volume (L). and reflex cough volume acceleration time.

Given that most PAS group differences were with induced‐reflex cough, we evaluated the associations between outcomes. Spearman's rho bivariate correlational tests revealed increased PAS scores during the 90 mL thin liquid boluses were significantly associated with reduced reflex cough volume acceleration (r = −0.629; P = 0.042) (Fig. 3A). Furthermore, reflex cough peak expiratory flow rise time was strongly correlated with the PAS scores in 90 mL; however, this positive association was not statistically significant (r = 0.593; P = 0.063) (Fig. 3B).

Figure 3.

Relationships between PAS scores during 90 mL thin liquid trials and (A) reflex cough volume acceleration, as well as (B) reflex cough peak expiratory flow rise time. Three participants had a PAS score of 2 and a rise time of .04 seconds; thus, there is only one data point is representing these individuals.

There was a limited range of DIGEST efficiency scores, such that most participants (n = 10) received scores of 0 or 1; this variable was excluded from analyses.

Discussion

Previous research repeatedly describes the association between swallow and cough deficits in PD, but to our knowledge this was the first study to examine this relationship of dysfunction in APD. The goals of this investigation were to elucidate the presence of concomitant swallow and cough deficits in APD and to examine whether disease duration has a significant relationship with swallow and cough dysfunction—important questions considering airway defense deficits contribute to respiratory infection, a leading cause of death in APD. ¹⁷ , ⁵⁶ , ⁵⁷ Although there was no association between primary swallow (PAS scores with thin liquid boluses) and cough (peak expiratory flow rate) outcomes, correlations between secondary outcomes indeed suggest that participants with abnormal airway invasion during swallowing have worse reflex cough function compared to individuals who had PAS scores of 1 or 2. Higher PAS scores were associated with worse reflexive cough outcomes, and disease duration was related to voluntary cough outcomes. Longer disease duration may result in greater disease severity that negatively impacts volitional cough control; however, different disease rating scales were used for each participant and we were unable to determine associations between disease severity and swallow/cough outcomes.

Informal reports of swallowing difficulty provided valuable information about individuals’ perception of impairments. The majority of participants reported perceived swallowing difficulty in their daily life and had impaired swallow safety upon instrumental swallowing exams, which differs from what is known about perception of airway protection in Parkinson's disease. For instance, perception of sensory input is often blunted in PD ¹ , ³ , ²⁷ , ³⁷ , ⁷⁰ ; thus, individuals may not perceive mechanical or chemical stimuli to the same degree as someone without PD. Despite APD affecting the basal ganglia, a region that largely regulates sensorimotor integration, ²² it may be that additional areas involved with sensory processing are differently impacted in APD versus PD. Or, due to rapid progression of APD, there may be less time to habituate to swallowing changes, resulting in intact perception of swallowing difficulty that is not detected during limited videofluoroscopic swallow studies, at least in the early stages of disease.

Although a small cohort, it was compelling that 82% either penetrated or aspirated during the 90 mL sequential drinking task. This swallow impairment co‐existed with reflex cough impairments that were characterized by significantly longer expiratory flow rise time and reduced inspiratory flow rate; and descriptively lower inspiratory volume and cough volume acceleration. Cough volume acceleration is a metric that quantifies cough efficiency, as it is a function of peak expiratory flow rate and peak expiratory flow rise time. Although the relationship between higher PAS scores and reflex cough expiratory flow rise time was not significant, rise times were longer than normal. Longer rise times will cause acceleration to decrease, even with adequate peak expiratory flow rates. These reflex cough volume acceleration deficits suggest that the respiratory pump was not strong or coordinated enough to quickly eject air in response to capsaicin. Previous work by Smith‐Hammond et al ⁴ and Pitts et al, ⁴⁰ , ⁴¹ found similar relationships in post‐stroke and PD cohorts, respectively. Strength and/or coordination of expiratory muscles for airway clearance seem to be impaired across many patient populations with swallow safety deficits. Not only do these impairments co‐exist with the presence of airway compromise, but they could compound adverse effects of aspiration since patients cannot adequately eject the material. Ideal cough volume acceleration may fluctuate, depending on what needs to be cleared; however, it reasons that high peak flow rates and low flow rise times result in fast and strong coughs for efficient and effective airway clearance. ⁴³ , ⁷¹ , ⁷²

Voluntary cough impairments are likely due to the degeneration of voluntary motor control in APD. ⁵⁵ , ⁷³ , ⁷⁴ , ⁷⁵ , ⁷⁶ Borders and Troche ⁷⁷ recently showed that peak expiratory flow rate during voluntary cough was associated with the percentage of aspirate material clearance in a heterogenous sample of neurodegenerative diseases. For instance, a peak expiratory flow rate of 3.0 L/s predicted ≥25% clearance. In the present study, the mean voluntary cough peak expiratory flow rate was only 1.4 L/s; thus, persons with APD may not adequately clear aspirate material. Although this study did not find any relationships between swallow dysfunction and low voluntary cough peak expiratory flow rate, there was larger cough expired volume for those with atypical PAS scores (≥3) and in those with longer disease duration.

As each form of APD progresses, it has greater impacts on the neural control of respiratory muscles needed for cough. Numerous muscles of the thorax (ie, internal intercostals, rectus abdominus, transverse abdominus, external and internal obliques), tongue (eg, genioglossus), pharynx (eg, pharyngeal constrictors), larynx (ie, thyroarytenoid, transverse arytenoids, cricothyroid), and hyoid are activated to maintain chest wall compression and high velocity airflow, particularly during a sequential cough. ¹² A sequential cough contains a patterned series of expulsive airflows associated with one inspiration, but separated by subsequent compression phases, referred to as cough reaccelerations. Each reacceleration repeatedly compresses the chest wall to reach low lung volumes, thereby increasing airflow and particulate removal from the lower, smaller airways. ⁴³ , ⁴⁵ In this study, most participants produced a sequential voluntary cough epoch, such that there was at least one cough reacceleration. Yet, a larger than normal volume of air was expired during the first cough in those with longer disease durations, suggesting a greater impairment in maintaining that systematic chest wall compression throughout sequential coughs. ⁴³ , ⁴⁵ , ⁷² Cough expired volume in the first cough may be different depending on cough inspired volume and how many coughs are produced ⁴⁵ ; however, atypical patterns of cough expired volume in sequential coughing are indicative of airway invasion in other patient populations.² Because those with PAS scores ≥3 had significantly larger voluntary cough expired volumes, it is possible that larger expired volumes would also be associated with more impaired swallowing, since swallowing does worsen with disease progression and severity. ¹ , ⁶⁰ For instance, Hegland and colleagues² noted that persons with PD and penetration or aspiration presented with atypical patterns of cough expired volume during sequential coughing. As such, they expired smaller or larger than normal volumes during the first cough of the epoch compared to historical controls and those with PD and no airway compromise.

Most participants presented with concomitant swallow and cough dysfunction, corroborating what others have found across patient populations. Associations between impaired cough airflows, increased PAS scores, and disease duration offer insight regarding which cough impairments may predict swallow dysfunction in the clinical setting and facilitate more efficient and timelier diagnostic and treatment protocols. Currently, clinicians have access to a peak flow meter—an accessible tool that measures voluntary peak expiratory flow rate. This tool can be immediately implemented into clinical practice to track changes in voluntary cough function, or use for cough rehabilitation. ⁷⁸ Although a peak flow meter is a pragmatic and currently applicable diagnostic/therapeutic tool, there remains a great need for clinical translation of laboratory‐based tools (ie, cough spirometry with pneumotachograph) to identify additional components of impaired cough, such as timing metrics and cough volumes—deficits that were significantly related to swallow safety and disease duration in this study. Not only do these findings support future research to clinically translate high‐tech cough assessment tools, but they also support investigations that evaluate which features of swallow and cough function are salient treatment targets for improved airway protection in APD, and other neurological diseases. Just as there are similar mechanisms of airway protective function, there may be similar mechanisms of dysfunction that could be simultaneously targeted in a comprehensive airway defense rehabilitation plan.

Limitations

There are several limitations to this study, which may be addressed in future work. First, only PAS and DIGEST efficiency scores for thin liquid boluses and self‐reported swallowing difficulty were used to quantify and describe the presence of swallow deficits, respectively. Nonetheless, these reports provided valuable insight into functional swallow deficits and how people with APD may differentially perceive swallowing deficits compared to those with PD. In the future, we plan to administer validated questionnaires that quantify patient reported outcomes. Furthermore, we were limited to videofluoroscopic swallow study reports. As such, we were not able to conduct reliability assessments on DIGEST or PAS scores; nor were we able to conduct timing and kinematic measures of swallowing in a variety of bolus types. Having this data would provide a more comprehensive assessment of relationships between objective swallow and cough outcomes.

Finally, the study was limited by a small sample size; an unfortunate consequence of recruiting participants who have extremely rare diseases. ¹⁵ Thus, there is limited generalizability from these findings.

Conclusion

Swallow safety impairments co‐exist, and are related to, reduced ability to effectively and efficiently clear the airways in forms of atypical Parkinsonsim. As disease duration increases, the ability to quickly expel air and maintain thoracic compression during sequential cough tasks decreases. Furthermore, impaired reflex cough (versus voluntary cough) may be more indicative of airway compromise during swallowing for individuals with APD. It is essential to evaluate cough and swallow throughout APD progression to initiate treatment plans that promote respiratory heath. However, future work is needed to clinically translate comprehensive cough assessment methods.

Author Roles

(1) Research Project: A. Conception, B. Organization, C. Execution; (2) Statistical Analysis: A. Design, B. Execution, C. Review and Critique; (3) Manuscript Preparation: A. Writing of the first draft, B. Review and Critique.

M.J.M.: 1A, 1B, 1C, 2A, 2B, 2C, 3A, 3B

J.C.: 1C, 2C, 3B

K.W.H.: 1A, 1B, 2A, 2C, 3B

Disclosures

Funding Sources and Conflicts of Interest: Funding sources include the Neuromuscular Plasticity Training Grant NIHT32 HD043730 (PI: D. Fuller) and NIH R01HD09165 (PI: K.Wheeler‐Hegland). The authors declare there are no conflicts of interest relevant to this work.

Financial Disclosures for the Previous 12 Months: During her predoctoral studies Michela J. Mir received funding from the NIHT32 HD043730 (PI: D. Fuller); she is currently receiving post‐doctoral funding from the BREATHE Training Grant, NIHT32 HL134621 (PI: G. Mitchell). Justin Childers received salary from the University of Florida and is currently the recipient of two medical school scholarships from FAU. Dr. Karen Wheeler Hegland receives funding from the NIH R01HD09165 and salary from the University of Florida.

Informed Consent

Verbal and written informed consent were obtained from all participants before enrolling in the study. All procedures were conducted in accordance to the Declaration of Helsinki and approved by the University’s Institutional Review Board (IRB202101196). We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this work is consistent with those guidelines.

Supporting information

Supplemental Table S1. Individual metrics for voluntary (VC) and reflex (RC) cough parameters. PIFR, peak inspiratory flow rate (L/s); IV, inspiratory volume (L); CPD, compression phase duration (s); PEFR, peak expiratory flow rate (L/s); PEFRT, peak expiratory flow rise time (s); CVA, cough volume acceleration (L/s/s); CEV, cough expired volume (L). and reflex cough volume acceleration time. Participant #6 was not able to produce a voluntary cough; thus, there are no VC values reported.