PAPER
Deglutitive laryngeal closure in stroke patients
M L Power, S Hamdy, S Singh, P J Tyrrell, I Turnbull, D G Thompson
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
See end of article for
authors’ affiliations
. . . . . . . . . . . . . . . . . . . . . . . .
Correspondence to:
Dr S Hamdy, Department of
Gastrointestinal Science
(Clinical Sciences Building),
Hope Hospital, Salford M6
8HD, UK; Shaheen.hamdy@
manchester.ac.uk
Received 14 July 2006
Revised 18 September 2006
Accepted
19 September 2006
Published Online First
29 September 2006
. . . . . . . . . . . . . . . . . . . . . . . .
J Neurol Neurosurg Psychiatry 2007;78:141–146. doi: 10.1136/jnnp.2006.101857
Background: Dysphagia has been reported in up to 70% of patients with stroke, predisposing them to
aspiration and pneumonia. Despite this, the mechanism for aspiration remains unclear.
Aims: To determine the relationship between bolus flow and laryngeal closure during swallowing in patients
with stroke and to examine the sensorimotor mechanisms leading to aspiration.
Methods: Measures of swallowing and bolus flow were taken from digital videofluoroscopic images in 90
patients with stroke and 50 healthy adults, after repeated volitional swallows of controlled volumes of thin
liquid. Aspiration was assessed using a validated Penetration–Aspiration Scale. Oral sensation was also
measured by electrical stimulation at the faucial pillars.
Results: After stroke, laryngeal ascent was delayed (mean (standard deviation (SD)) 0.31 (0.06) s,
p,0.001), resulting in prolongation of pharyngeal transit time (1.17 (0.07) s, p,0.001) without a
concomitant increase in laryngeal closure duration (0.84 (0.04) s, p = 0.9). The delay in laryngeal elevation
correlated with both the severity of aspiration (r = 0.5, p,0.001) and oral sensation (r = 0.5, p,0.001).
Conclusions: After stroke, duration of laryngeal delay and degree of sensory deficit are associated with the
severity of aspiration. These findings indicate a role for sensorimotor interactions in control of swallowing and
have implications for the assessment and management of dysphagia after stroke.
Dysphagia has been reported in up to 70% of patients withstroke, with the major clinical complication beingdeglutitive aspiration.1 Historically, the prevailing belief
has been that for most patients, aspiration resolves sponta-
neously within days.2 In fact, aspiration is now known to persist
well beyond 6 months in up to 50% of patients with stroke,3
and the incidence of pneumonia remains unacceptably high.4
Despite this, objective measurements of relationships between
bolus flow and laryngeal closure, factors that might predict
patients at risk of aspiration, have never been systematically
determined in a cohort of patients with pure stroke.
Patients with hemispheric stroke are described as presenting
with ‘‘difficulty initiating coordinated movement’’,5 ‘‘delayed
pharyngeal response’’6 and ‘‘delayed or absent swallow reflex’’.7
It has been speculated that these features are associated with a
breakdown in the descending efferent outputs from the cortex,
which initiate the brain stem-mediated pharyngeal response
and laryngeal protection.8 However, the neural control of
swallowing integrates peripheral afferent inputs from the
oropharynx with descending efferent outputs.9 Recent work
by Aviv et al10 has suggested that patients with stroke and
dysphagia also have impairments in pharyngeal sensation.
Despite these observations, whether such sensory deficits
specifically affect deglutitive laryngeal function remains
unknown.
The protective mechanism that guards against aspiration is
laryngeal closure, which is achieved by a sequence of events,
beginning with cessation of respiration, approximation of the
arytenoids and adduction of the vocal cords. Complete closure
is achieved by a combination of laryngeal ascent and epiglottic
descent.11 12 Laryngeal closure duration (LCD) modulates to
accommodate bolus characteristics13; however, whether the
timing of laryngeal ascent is fixed or responsive to bolus
characteristics is not known.14 15
Studies on swallowing in humans have implicated the faucial
pillar, an area innervated by the glossopharyngeal nerve, as a
‘‘trigger’’ for the onset of laryngeal ascent.14 16 17 Studies on
patients with aspirating stroke have suggested that delayed
onset of laryngeal ascent can be shortened by stimulation at the
faucial pillar.18 Stimulation at the faucial pillar has been
considered beneficial because it stimulates mechanoreceptors
at the head of the hypopharynx, which relay information to the
brain stem and cortex to initiate laryngeal closure. The evidence
to support this contention is, however, controversial, with some
authors suggesting that higher cortical functions are more
important.19 Furthermore, some studies have suggested that the
important receptive sensory fields for triggering swallowing do
not include the faucial pillar region.20 Thus, although the faucial
pillar is the anatomical gateway to the hypopharynx, it may
have no role in the onset of laryngeal ascent. A systematic
examination of this relationship in health and stroke is,
however, lacking.
The aims of this study were therefore to examine the
temporal linkage of bolus transit and laryngeal closure in
healthy adults and to compare our findings with those obtained
from patients after hemispheric stroke. We hypothesised that in
healthy participants, both the initiation and duration of
laryngeal closure would be modulated by sensory input, and
that this relationship would be impaired in patients with
stroke. We also postulated that altered oral sensation would
relate to the severity of aspiration in patients with stroke.
METHODS
Participants
In all, 140 people took part in the studies (50 healthy
participants, 29 men, mean age 38 years, in two age groups
,55 (n = 38) and .55 years (n = 11) and 90 participants with
hemispheric stroke, 53 men, mean age 69 years, range 29–
92 years). Healthy participants were recruited from the local
community. Patients with stroke were recruited from a large
teaching hospital and studied within 2 weeks of stroke (mean
interval 10 days, range 6–13 days). Participants were excluded
if they had a history of difficulty in swallowing, neurological
disease, intercurrent illness or upper gastrointestinal disease.
Abbreviations: LCD, laryngeal closure duration; OLE, onset of laryngeal
elevation; OTT, oral transit time; PAS, Penetration–Aspiration Scale; PTT,
pharyngeal transit time; SRT, swallow response time
141
www.jnnp.com

They were also ineligible if they were unable either to give
informed consent or unable to participate in the videofluoro-
scopy examination. All study protocols were presented to and
approved by the local research ethics committee, and all
participants gave written consent before the study began.
Assessment of damage caused by stroke
In all patients, computerised tomography scans of the brain were
used to confirm the type, size and site of the lesion. All scans
were analysed by a neuroradiologist, blinded to the patient’s
clinical presentation and swallowing status. Lesions were
characterised according to their type (ischaemic v haemorrha-
gic), side (left or right) and volume, using the Alberta Stroke
Programme Early CT Score.21 The 100-point Barthel Index was
recorded as an indication of functional impairment.22
Assessment of swallowing
All participants were assessed by videofluoroscopy using barium
liquid (60% wt/vol, EZ-HD, E-Z-EM, London, UK). Images were
acquired (Siemens Fluorospot H SIRESKOP SX Unit, Siemens
Aktiengesellschaft Medical Engineering, Erlangen, Germany) in
real time using continuous fluoroscopy at 30 frames/s (Videomed
DI TV system) and recorded by a digital video at 25 frames/s (Sony
DHR 1000, Sony UK, Weybridge, Surrey, UK). Lateral images of
the oropharynx were subsequently acquired without magnifica-
tion, according to previously described protocols.23 The total
screening time was kept below 80 s (range 42–73 s) in all cases,
giving a radiation dose of ,0.3 mSv.
Videofluoroscopic measures
Two measures of bolus transit were chosen to evaluate bolus
flow through the mouth: oral transit time (OTT) and
pharyngeal transit time (PTT). In addition, measures of time
from arrival of the bolus head at the hypopharynx (swallow
response time, SRT) to onset of laryngeal elevation (OLE) and
duration of laryngeal closure (LCD) were recorded.
Definition of terms
OTT was defined as the interval, in seconds, between the first
frame showing elevation of the tongue tip (with subsequent
posterior movement of the bolus tail) and the first frame
showing the arrival of the head of the bolus at the hypopharynx
(defined as the anatomical point where the ramus of the
mandible crosses the tongue base).
PTT was defined as the interval, in seconds, between the first
frame showing the arrival of the bolus head at the hypopharynx
to the last frame showing the tail of the bolus passing through
the mid-margin of the upper oesophageal sphincter.
SRT was defined as the interval, in seconds, between the first
frame showing the arrival of the bolus head at the hypopharynx
to the first frame showing upward excursion of the larynx.
LCD was defined as the interval, in seconds, between the first
frame showing contact between the inferior surface of the
epiglottis and arytenoids and the first frame showing that
contact had ceased.
Temporal reference points
In addition to the primary measures of bolus transit and
laryngeal closure, in all healthy participants and 52 patients
with stroke, the temporal relationship between the elevation of
the tongue tip (swallow onset) and laryngeal closure was
obtained by referencing three additional time points: the OLE,
the time to complete laryngeal elevation and the time to
laryngeal reopening, with the first frame denoting the onset of
elevation of the tongue tip (time = 0).
Laryngeal penetration and aspiration were assessed for all
swallows using a previously developed and validated 8-point
Penetration–Aspiration Scale (PAS), which describes the
severity of airway compromise.24 Given that normal participants
are known to score 1–2 on the PAS,25 patients were considered
to have abnormal laryngeal protection (aspirators) if they
scored >3 on one or more swallows on the PAS. Their worst
aspiration score was used to determine the relationship with
demographic and swallowing measures.
Assessment of sensation of the faucial pil lar
To provide an objective and reliable measure of oral sensation,
the sensitivity of the faucial pillar was assessed using a 2-mm
fingertip electrode (St Mark’s Pudendal Electrode, Medtronic
Diagnostics A/S, Tonsbakken, Skovlunde, Denmark) placed
digitally on to each anterior faucial pillar. The electrode was
connected to a constant current stimulator (Model DS7,
Digitimer, Welwyn Garden City, Herts, UK), and the stimuli
were delivered via a trigger generator (Model DG2, Digitimer)
at a frequency of 5 Hz (square wave duration 200 ms). For each
faucial pillar, the intensity of the stimulus was increased in a
stepwise fashion in increments of 0.2 mA from zero until the
participant just perceived the electrical sensation. This process
was repeated, across the right and left faucial pillars, three
stimulations per side, in random order.
EXPERIMENTAL PROTOCOLS
Protocol 1: Pharyngeal response patterns in health
In all, 38 healthy participants were assessed by videofluoro-
scopy using liquid boluses (665 ml). All participants were
asked to retain the liquid in their mouth until instructed to
swallow.
Protocol 2: Sensory thresholds and the effect of bolus
volume in health
Sensation of the faucial pillar was measured in 12 healthy
participants (6 men, mean age 38 years, range 28–60 years)
after protocol 1 to determine the normal range of sensation of
the faucial pillar.
A further 12 healthy participants (10 men, mean age
39 years, range 24–55 years) were also assessed using boluses
of 5, 10, 15 and 20 ml of barium to determine the effect of bolus
size on the initiation and duration of laryngeal closure.
Protocol 3: Pharyngeal response after cortical
hemispheric stroke
In addition to the measures of bolus flow and laryngeal closure
described above, airway compromise was evaluated for all
swallows using the PAS for the 90 stroke patients.19
Protocol 4: Sensory thresholds and the effect of bolus
volume in stroke
Sensation of the faucial pillar was measured in 41 patients
recruited to protocol 3.
Ten patients from protocol 3 were assessed using multiple
bolus volumes of 5, 10, 15 and 20 ml of barium to determine
the effect of bolus size on the initiation and duration of
laryngeal closure.
DATA ANALYSIS
All data are expressed as mean (standard error of mean). A p
value ,0.05 was taken to indicate significance unless otherwise
stated. Descriptive and comparative analyses were made using
SPSS software V.11.1. Comparison of the performance within
and between participants was made using analysis of variance.
Interparticipant comparisons were made using individual
means for each measure. Spearman’s correlation coefficients
were used to explore the relationships between age, video-
fluoroscopy timings and severity of aspiration.
142 Power, Hamdy, Singh, et al
www.jnnp.com