Oropharyngeal Dysphagia after Stroke: Incidence, Diagnosis,

and Clinical Predictors in Patients Admitted to

a Neurorehabilitation Unit

Paolo Falsetti, MD, PhD,* Caterina Acciai, MD, PhD,* Rosanna Palilla, MD,*
Marco Bosi, MD,† Francesco Carpinteri, MD,* Alberto Zingarelli, MD,†
Claudio Pedace, MD,‡ and Lucia Lenzi, MD*

Objective: We analyzed patients with stroke in a neurorehabilitation unit to define

incidence of dysphagia, compare clinical bedside assessment and videofluoroscopy
(VFS), and define any correlation between dysphagia and clinical characteristic of
patients. Methods: In all, 151 consecutive inpatients with recent ischemic or hemor-
rhagic stroke were enrolled. Results: Dysphagia was clinically diagnosed in 62 of 151
patients (41%). A total of 49 patients (79% of clinically dysphagic patients) under-
went VFS. Six patients clinically suggested to be dysphagic had a normal VFS find-
ing. The correlation between clinical and VFS diagnosis of dysphagia was significant (r
5 0.6505). Penetrations and aspirations were observed, respectively, in 42.8% and
26.5% of patients with dysphagia, with 12.2% classified as silent. Lower respiratory
tract infections were observed in 5.9%, more frequently in patients with dysphagia
(30%). Dysphagia was not influenced by type of stroke. Cortical stroke of nondominant
side was associated with dysphagia (P 5 .0322) and subcortical nondominant stroke
showed a reduced frequency of dysphagia (P 5 .0008). Previous cerebrovascular dis-
ease resulted associated to dysphagia (P 5 .0399). Patients with dysphagia had signif-
icantly lower functional independence measurement (FIM) and level of cognitive
functioning on admission and lower FIM on discharge, longer hospital stay, and
more frequent malnutrition, and they were more frequently aphasic and dysarthric.
Percutaneous endoscopic gastrostomy was used in 18 of 151 patients (11.9%) (41.8%
of patients with VFS-proved dysphagia). Conclusions: Dysphagia occurs in more
than a third of patients with stroke admitted to rehabilitation. Clinical assessment
demonstrates good correlation with VFS. The grade of dysphagia correlates with
dysarthria, aphasia, low FIM, and level of cognitive functioning. Large cortical strokes
of nondominant side are associated with dysphagia. Key Words: Dysphagia—
stroke—videofluoroscopy—percutaneous endoscopic gastrostomy.
Ó 2009 by National Stroke Association

Dysphagia is a disorder of deglutition affecting the oral, ing physiology of the upper aerodigestive tract and it oc-
pharyngeal, and/or esophageal phases of swallowing. curs frequently after stroke, with an incidence ranging
Oropharyngeal dysphagia is any abnormality in swallow- widely between 29% and 81%.1-3 This discrepancy be-
tween studies depends on different methods of diagnosis,
From the *Neurorehabilitation, †Radiology; and ‡Internal Medicine time after stroke, and types of lesion. Aspiration (passage
and Geriatrics Units, Local Health Unit 8, S. Donato Hospital, Arezzo,
Italy.
of material into the larynx below the true vocal cords) af-
Received January 1, 2009; accepted January 13, 2009. ter swallowing, especially fluids, is probably the most se-
Address correspondence to Paolo Falsetti, MD, PhD, Neurorehabi- vere aspect of oropharyngeal dysphagia with incidence
litation Unit, Local Health Unit 8, S. Donato Hospital, Arezzo, Italy. between 22% and 52%.4-6 Nearly half of aspirations in pa-
E-mail: [email protected]. tients with stroke are silent7,8 and they have been associ-
1052-3057/$—see front matter
Ó 2009 by National Stroke Association
ated with increased morbidity and mortality in many
doi:10.1016/j.jstrokecerebrovasdis.2009.01.009 studies.9

Journal of Stroke and Cerebrovascular Diseases, Vol. 18, No. 5 (September-October), 2009: pp 329-335 329
330 P. FALSETTI ET AL.

Moreover, the presence of oropharyngeal dysphagia in stroke) was recorded and classified (no cerebrovascular
patients recovering from stroke has often been associated disease, diffuse cerebrovascular disease, previous stroke).
with malnutrition, dehydration, pulmonary infections, Classification of stroke location was made on the basis of
prolonged hospital stay, and death.10,11 Lower respiratory the damaged area on CT/magnetic resonance imaging in 7
tract infection is probably the most severe event related to subtypes: cortical stroke on dominant side (generally left);
dysphagia in the early period after stroke, and it is more cortical stroke on nondominant side (generally right);
common in patients with oropharyngeal dysphagia and subcortical stroke on dominant side (generally left);
aspirations.12 subcortical stroke on nondominant side (generally right);
The early diagnosis of aspiration should induce clini- brainstem stroke; cerebellar stroke; and mixed/multifocal
cians to limit oral administration of nutrients or drugs stroke. This classification derived from previous radio-
to reduce the incidence of pulmonary infections. How- logic and clinical methods of ischemic stroke classifica-
ever, bedside swallowing assessments lack the accuracy tion19,20 with indication of side location.21 In particular,
to be used as a screening test in stroke, in particular in pa- cortical stroke corresponds to total or partial anterior circu-
tients with alteration of consciousness.7,13 Moreover, bed- lation stroke of Oxfordshire Community Stroke Project–
side prediction of aspiration seems to be inaccurate.8,14,15 derived CT stroke classification, subcortical stroke
Videofluoroscopy (VFS) can be considered as gold stan- corresponds to small partial anterior circulation stroke
dard in diagnosing dysphagia with aspirations (silent or (striatocapsular) or lacunar stroke, and brainstem or cere-
not) because of the capability to study the entire process bellar stroke corresponds to posterior circulation stroke.19
of deglutiton.16-18 Nevertheless, this examination necessi- The patients were enrolled regardless of cognitive func-
tates patient collaboration and sitting posture, so it cannot tion, consciousness level, grade of collaboration, or capa-
be proposed for all patients in the very early period after bility of communication. Presence of aphasia (defined as
stroke. deficit in at least one of the 4 language areas: comprehen-
In this study we have undertaken a prospective analy- sion, fluency, naming, and repetition) and dysarthria
sis of consecutive patients with stroke in a postintensive (defined as deficit of articulatory agility of speech) was
neurorehabilitation unit to define incidence of oropharyn- recorded for each patient.21
geal dysphagia (both by clinical and instrumental On admission and on discharge each patient underwent
methods), compare clinical bedside assessment and VFS neurologic and functional assessment, even with defini-
(considered as gold standard), and define any correlation tion of functional independence measurement (FIM) score
between the presence of dysphagia (both clinically and and level of cognitive functioning (LCF) score.
VFS proved) and clinical characteristic of patients with Nutritional state was assessed by clinical and biochem-
stroke (nutritional and functional status, type of stroke, ical parameters on admission and on discharge determin-
incidence of pulmonary infections). ing serum levels of albumin, ferritin, iron, urea, and
lymphocyte count, and weight loss. Malnutrition was
diagnosed when at least 2 of 5 parameters were altered.22
Methods
Lower respiratory tract infections (diagnosed with
In all, 151 consecutive inpatients admitted to our neuro- fever .38 C and abnormal chest radiograph result)
rehabilitation unit between January 2005 and December from the time of stroke to discharge were recorded.3,6
2006 with diagnosis of previous ischemic or hemorrhagic A standardized clinical bedside test was performed by
stroke were enrolled in this study. All patients were trans- the doctor in each patient within 1 day from admission.
ferred to the neurorehabilitation unit after the acute This test was arranged in 3 steps and it was influenced
phase, with a mean length of stay on intensive care unit by previous works6,17,23 and guidelines.10
of 13 days (range 6-21). The first step served to identify the level of conscious-
Patients with a history of head and neck damage, his- ness and collaboration of patient (patients with LCF ,4
tory of neurologic disease other than cerebrovascular dis- were immediately considered dysphagic) and to define
orders, or current dysphagia were excluded from the oral motor and sensory assessment (voice quality; speech
study. The ethical committee of the hospital approved and language; swallowing of saliva; movements of cricoid
the study. cartilage; lips, tongue, and velopharynx; gag reflex; pres-
The characteristics of study participants were: 77 male, ervation of pharyngeal sensation; capability of voluntary
74 female, mean age 79.4 years (range 58-91), and mean cough).
duration of disease (time from stroke) 14 days. The second step comprised the swallowing of 5 mL of
The diagnosis of stroke was always confirmed by com- water with concomitant pulse oxymetry, carefully observ-
puted tomography (CT), magnetic resonance imaging, or ing signs of oral-facial apraxia (loosening of water from
both, and the last scan of each patient was reviewed to lips, delay in swallowing, abnormality or absence of
classify the lesion location (Table 1) and to define the tongue movements) or signs of penetration/aspiration
type of stroke (ischemic or hemorrhagic). The presence (‘‘wet’’ or ‘‘gurgly’’ voice, coughing, .2% decrease of
of signs of cerebrovascular disease (other than the recent basal value of oxygen saturation at pulse oximetry).
DYSPHAGIA FOLLOWING STROKE 331

Table 1. Demographic and clinical characteristics of patients

Overall study population No dysphagia Dysphagia P value (statistic test)

No. of patients (n) 151 89 (58.9%) 62 (41.0%)

Female (n) 74 43 31 ns (Chi square)
Male (n) 77 46 31 ns (Chi square)
Mean age, y (range; SD) 79.4 (58-91; 6.2) 78.6 (58-87; 6.6) 80.7 (58-91; 5.4) ns (Mann-Whitney)
Mean duration of disease 14 13 15 ns (Mann-Whitney)
(days from stroke)
Type of recent stroke
 Ischemic 112 (74.1%) 65 (73.0%) 47 (75.8%) ns (Chi square)
 Hemorrhagic 39 (25.8%) 24 (26.9%) 15 (24.1%)
Classification of stroke .0251 overall (Chi square)
location
 Cortical stroke- 33 (21.8%) 18 (19.7%) 15 (25%) asp: 3 ns (Fisher)
dominant side (generally
left) TACI or PACI
 Cortical stroke- 35 (23.1%) 15 (16.8%) 20 (32.2%) asp: 5 .0322 (Fisher)
nondominant side
(generally right) TACI or
PACI
 Subcortical stroke- 24 (15.8%) 12 (7.9%) 12 (20%) asp: 4 ns (Fisher)
dominant side (generally
left) striatocapsular-
PACI or LACI
 Subcortical stroke- 30 (19.8%) 26 (28.5%) 4 (6.6%) asp: 0 .0008 (Fisher)
nondominant side
(generally right)
striatocapsular-PACI or
LACI
 Brainstem stroke POCI 16 (10.5%) 8 (8.7%) 8 (13.3%) asp: 0 ns (Fisher)
 Cerebellar stroke POCI 11 (7.2%) 8 (8.7%) 3 (5%) asp: 1 ns (Fisher)
 Mixed or multifocal 2 (1.3%) 2 (2.1%) 0% asp: 0 ns (Fisher)
stroke
Cerebrovascular disease
 No cerebrovascular 17 (11.2%) 14 (15.3%) 3 (5%) .0399 (Fisher: disease v no
disease disease)
 Diffuse cerebrovascular 110 (72.8%) 66 (74.1%) 44 (70.9%)
disease
 Previous stroke 24 (15.8%) 9 (9.8%) 15 (25%) ns (Fisher: diffuse v focal)
LCF at admission (range; 6.3 (3-8; 1.3) 6.7 (4-8; 1.0) 5.7 (3-8; 1.4) ,.0001 (Mann-Whitney)
SD)
FIM (range; SD)
 At admission 45.9 (18-106; 21.6) 53.1 (18-106; 20.1) 35.6 (18-92; 19.6) ,.0001 (Mann-Whitney)
 At discharge 59.6 (18-129; 26.9) 68.5 (19-129; 25) 46.8 (18-122; 24.4) ,.0001 (Mann-Whitney)
Length of stay, days (range; 29.9 (6-93; 14.7) 26.6 (6-60; 12.3) 35 (13-93; 16.7) .0012 (Mann-Whitney)
SD)
Lower respiratory tract 9 (5.9%) 1 (1.1%) 8 (12.9%) .0036 (Fisher)
infection
Malnutrition 132 (87.4%) 73 (82.0%) 59 (95.1%) .0230 (Fisher)
Aphasia 48 (31.7%) 16 (17.5%) 32 (53.3%) ,.0001 (Fisher)
Dysarthria 83 (54.9%) 28 (30.7%) 55 (91.6%) ,.0001 (Fisher)
PEG 18 (11.9%) 0% 18 (30%) ,.0001 (Fisher)

Abbreviations: asp, aspirating at videofluoroscopy; FIM, functional independence measurement; LACI, lacunar infarction; LCF, level of cog-
nitive functioning; ns, not significant; PACI, partial anterior circulation infarction; PEG, percutaneous endoscopic gastrostomy; POCI, posterior
circulation infarction; TACI, total anterior circulation infarction.
332 P. FALSETTI ET AL.

The third step consisted of swallowing at least 20 mL of

water, with the same procedures just described.
On the basis of the results of the 3-step bedside swal-
lowing assessment, each patient was described as dys-
phagic (abnormality of at least one of the items
described in each step) or nondysphagic (complete nor-
mality in each step). Clinically proved dysphagia was
classified in 3 categories (1 5 dysphagia prevalently af-
fecting oral phase, 2 5 dysphagia prevalently affecting
pharyngeal phase, 3 5 mixed dysphagia).
Patients classified as clinically dysphagic were further
evaluated with VFS. VFS was completed with a standard
protocol4,8,24 within the first week of admission. A video-
recording of the oral cavity and pharynx was obtained in
the lateral plane while the patient swallowed in sequence
5 mL and 10 mL of solution of barium of different consis-
tency (liquid at later attempts of the examination). The ex-
aminations were stopped if the patient exhibited
significant aspiration.
VFS-proved dysphagia was classified into 4 categories
(0 5 no dysphagia, 1 5 dysphagia affecting oral phase,
2 5 dysphagia affecting pharyngeal phase, 3 5 mixed
dysphagia). The ordinal penetration-aspiration scale
was used for scoring airway invasion (Figs 1 and 2).25
Statistical analysis was performed with software (InStat 3 Figure 1. Female, 89 years. Patient with left cortical ischemic stroke and
GraphPad, GraphPad Software Inc., LaJolla, CA 92037). The diffuse cerebrovascular disease, right hemiplegia, nonfluent aphasia, apraxia,
Chi-square test was used for an overall approach to compare oropharyngeal dysphagia. VFS demonstrated grade 5 on penetration-aspira-
tion scale (contrast agent contacting vocal folds with visible residue and with-
the percentages among groups. Fisher exact test was used to
out patient response). Decreased capability of oral clearance. Lateral scan.
compare the percentages between two groups. Pearson sta-
tistics and Spearman rank test were applied to correlate vari-
ables (respectively with or without normal distribution). In all, 49 of 151 patients (79% of clinically dysphagic
The level of statistical significance was set at a P level of .05. patients) underwent VFS. Dysphagia prevalently affecting
oral phase was observed in 15, dysphagia prevalently af-
fecting pharyngeal phase was observed in 13, and mixed
Results
form of dysphagia was observed in 15. Six patients clini-
The patients were admitted with a mean LCF of 6.31 cally suggested to be dysphagic had a normal VFS result.
(range 3-8) and FIM of 45.9 (range 18-106). Mean FIM at Spearman correlation between clinical and VFS diagnosis
discharge was 59.6 (range 18-129), with a mean increment of dysphagia was highly significant (r 5 0.6505, P ,.0001).
of 13.7. The mean length of stay on neurorehabilitation Penetrations were observed in 21 of 49 patients (42.8%).
unit was 29.9 days (range 6-93). Aspirations were observed in 13 of 49 patients (26.5%)
Imaging techniques demonstrated no previous cerebro- and in 6 cases (12.2%) were classified as silent and signif-
vascular disease in 17 of 151 patients, diffuse cerebrovas- icant.
cular disease in 110 of 151, and a previous stroke in 24 of The diagnosis of dysphagia was not influenced by type
151. The prevalence of lesion location of the recent stroke of stroke. Instead, stroke location and side showed signif-
is reported in Table 1. The type of stroke was ischemic in icant association with dysphagia (Table 1). In particular,
112 patients and hemorrhagic in 39. cortical stroke of nondominant side showed more fre-
Aphasia was observed in 48 patients (31.7%) and dysar- quent dysphagia (P 5 .0322). On the other hand, subcorti-
thria in 83 (54.9%). Biochemical diagnosis of malnutrition cal nondominant stroke showed a reduced frequency of
was made in 132 patients (87.4%). dysphagia (P 5 .0008). In addition, the presence of previ-
Dysphagia was clinically diagnosed in 62 of 151 pa- ous cerebrovascular damage (indifferent if focal or dif-
tients (41%). Five patients (3.3%) were considered dys- fuse) was associated with dysphagia (P 5 .0399).
phagic without clinical testing because of their low Lower respiratory tract infections until discharge were
awareness level (LCF , 4) on admission. observed in 9 of 151 patients (5.9%) and they were signif-
Dysphagia prevalently affecting oral phase was ob- icantly more frequent in those with dysphagia (P 5.0036).
served in 23 patients, dysphagia prevalently affecting pha- Dysphagic patients also had lower LCF on admission
ryngeal phase in 14, and mixed form of dysphagia in 25. (P , .0001) and discharge (P , .0001), longer hospital
DYSPHAGIA FOLLOWING STROKE 333
26 2
between 25% and 81%. These prevalences are probably
dependent on which method is used for the diagnosis, the
time from stroke, and the severity of stroke in the case
study.
The results of this study have been derived from a hos-
pital-referred cohort of consecutive patients with stroke
admitted to a neurorehabilitation unit. The timing of the
assessment and the diagnostic methods could have deter-
mined some bias in the results with respect to similar
studies.
Our results confirm that dysphagia occurs in more than
a third (41%) of consecutive patients with stroke admitted
to a neurorehabilitation unit. Feeding gastrostomy tubes
(PEG) were inserted in 11.9% of our neurorehabilitation
population. The percentage of severe dysphagic stroke re-
quiring PEG is comparable with previous neurorehabili-
tation studies.27 A direct correlation was observed
between insertion of PEG and aphasia, aspiration at
VFS, and length of rehabilitation hospital stay, and an in-
verse correlation with FIM and LCF at admission, and im-
provement of FIM score at discharge. A previous
rehabilitation study confirmed significant lower FIM
scores in patients with stroke who require PEG placement
for the management of dysphagia and not significant dif-
Figure 2. Female, 86 years. Patient with right cortical ischemic stroke, left ferences in FIM improvement and length of hospital stay
hemiplegia, hemispatial neglect, and oropharyngeal dysphagia. VFS demon- in these patients.28
strated grade 8 on penetration-aspiration scale (contrast agent passing glot- In our study 79% of clinically dysphagic patients under-
tis; visible subglottic residue without patient response). Anteroposterior
went VFS. Six patients clinically suggested to be dysphagic
scan.
had a normal VFS finding. However, the clinical classifica-
tion of dysphagia showed good correlation with VFS.
stay (P 5 .0012), and more frequent malnutrition (P 5 A comparison with the previous study of Smithard
.023), and they were more frequently aphasic and dys- et al13 in patients with acute stroke shows a low sensibility
arthric (P , .0001). (47%) of bedside assessment in predicting aspiration.
Spearman analysis showed significant direct correlation However, in our study the diagnosis of dysphagia com-
between the grade of clinically proved dysphagia and dys- prised even oral phase alterations, and not only aspira-
arthria (r 5 0.2962, P 5 .0002), aphasia (r 5 0.3553, P , tions. This could probably increase the prevalence of
.0001), and inverse correlation with FIM (r 5 –0.2925, P clinical dysphagia and subsequently improve the sensi-
5 .0287) and LCF (r 5 –0.3383, P 5 .0082) at admission. tivity of bedside assessment with regard to VFS.
Percutaneous endoscopic gastrostomy (PEG) was used Therefore, clinical bedside assessment is a necessary
in 18 of 151 patients (11.9%; 41.8% of 43 patients with VFS- screening test, with high correlation with VFS. However,
proved dysphagia). criteria that is too wide for clinical diagnosis of dysphagia
Correlation analysis showed significant direct correla- could generate false-positive findings, as in our case
tion between PEG and grade of aspiration at VFS (r 5 study.
0.4339, P 5 .0015), aphasia (r 5 0.5482, P , .0001), and Lower respiratory tract infections were observed in
length of rehabilitation hospital stay (r 5 0.3189, P 5 5.9% of patients. The pneumonia incidence in our study
.0115), and inverse correlation with FIM (r 5 –0.4969, is lower than in previous rehabilitation studies, where it
P , .0001) and LCF (r 5 –0.4893, P , .0001) at admission ranged widely between 7% and 29%.26,29,30 Nevertheless,
and improvement of FIM score at discharge (r 5 –0.3622, in patients with dysphagia, pneumonia was diagnosed
P 5 .0061). with significantly higher frequency.
A direct correlation was observed between grade of
dysphagia and dysarthria and aphasia, and inverse corre-
Discussion
lation was observed with FIM and LCF at admission.
Dysphagia is a common complication after stroke, but Moreover, in the dysphagic group, significantly longer
its prevalence is widely discrepant in the various studies, length of stay, frequency of malnutrition, and lower FIM
ranging between 29%1 and 81%.2,3 The frequency of and LCF have been observed. These correlations were
dysphagia in rehabilitation studies ranges widely observed in previous work.10,27
334 P. FALSETTI ET AL.

The association between aphasia and dysphagia was 2. Meng NH, Wang TG, Lien IN. Dysphagia in patients with
observed by Barer,1 whereas in a more recent study brainstem stroke: Incidence and outcome. Am J Phys
Schroeder et al21 underlined that hemispatial neglect Med Rehabil 2000;79:170-175.
3. Martino R, Foley N, Bhogal S, et al. Dysphagia after
shows a stronger association with dysphagia than apha- stroke: Incidence, diagnosis, and pulmonary complica-
sia. In our study we did not consider hemispatial neglect tions. Stroke 2005;36:2756-2763.
as an independent variable. 4. Kidd D, Lawson J, Nesbitt R, et al. Aspiration in acute
However, in our case study cortical stroke of nondom- stroke: A clinical study with videofluoroscopy. QJM
inant side showed significantly higher prevalence of dys- 1993;86:825-829.
5. Lim SH, Lieu PK, Phua SY, et al. Accuracy of bedside clin-
phagia, in contrast to subcortical nondominant stroke that ical methods compared with fiberoptic endoscopic exam-
was significantly not associated with dysphagia. As hemi- ination of swallowing (FEES) in determining the risk of
spatial neglect is characteristically associated to cortical aspiration in acute stroke patients. Dysphagia 2001;
right (nondominant) hemispheric damage but not associ- 16:1-6.
ated to internal capsule (subcortical) and lacunar right- 6. Mann G, Hankey GJ, Cameron D. Swallowing function
after stroke: Prognosis and prognostic factors at 6
sided damage,21,31 our results indirectly confirm those months. Stroke 1999;30:744-748.
of Schroeder et al.21 Other studies confirmed that right 7. Terré R, Mearin F. Oropharyngeal dysphagia after the
hemisphere lesions are associated to dysphagia with aspi- acute phase of stroke: Predictors of aspiration. Neurogas-
rations.32,33 troenterol Motil 2006;18:200-205.
No other correlation was observed in our study be- 8. Splaingard ML, Hutchins B, Sulton LD, et al. Aspirations
in rehabilitation patients: Videofluoroscopy vs bedside
tween dysphagia and location or type of stroke. This find- clinical assessment. Arch Phys Med Rehabil 1988;
ing was consistent with results from previous studies,21,34 69:637-640.
and our lesion mapping in patients with stroke undergo- 9. Ramsey D, Smithard D, Kalra L. Silent aspiration: What
ing rehabilitation has not indicated differences in swal- do we know? Dysphagia 2005;20:218-225.
lowing dysfunction (oral dysmotility, pharyngeal 10. Martino R. Preferred practice guideline for dysphagia
2000. Available from: URL:http://www.caslpo.com/
dysfunction, penetration, aspiration) with different loca- english_site/DYSPHAGIADOCUMENT.pdf. Accessed
tion or type of lesion. February 14, 2007.
Several studies indicated that large cortical hemi- 11. Crary MA, Carnaby-Mann GD, Miller L, et al. Dysphagia
spheric lesions (with or without indication of side) dem- and nutritional status at the time of hospital admission
onstrate the highest incidence of dysphagia.33,35-38 Our for ischemic stroke. J Stroke Cerebrovasc Dis 2006;
15:164-171.
results demonstrate significantly higher incidence of dys- 12. Kidd D, Lawson J, Nesbitt R, et al. The natural history
phagia only for right cortical strokes. However, the corre- and clinical consequences of aspiration in acute stroke.
lations between dysphagia and aphasia and dysarthria QJM 1995;88:409-413.
indirectly confirm the importance of left cortical hemi- 13. Smithard DG, O’Neill PA, Park C, et al. Can bedside as-
spheric lesions. The correlation with low scores of FIM sessment reliably exclude aspiration following acute
stroke? Age Ageing 1998;27:99-106.
and LCF could indirectly indicate that large and more se- 14. Linden P, Siebens AA. Dysphagia: Predicting laryngeal
vere lesions can increase the incidence of dysphagia.36,37 penetration. Arch Phys Med Rehabil 1983;64:281-284.
Posterior circulation strokes were also not associated 15. Linden P, Kuhlemeyer KV, Paterson C. The probability of
with dysphagia in our study. This finding is incongruent correctly predicting subglottic penetration from clinical
with previous studies that identified severe dysphagia in observations. Dysphagia 1993;8:170-179.
16. Horner J, Massey EW. Silent aspiration following stroke.
patients with stroke and brainstem lesions.2,30,39 On the Neurology 1988;38:317-319.
other hand, many studies confirmed that brainstem and 17. Smith HA, Lee SH, O’Neill PA, et al. The combination of
cerebellar stroke are not significantly associated with bedside swallowing assessment and oxygen saturation
higher prevalence of dysphagia.35-37,40-42 monitoring of swallowing in acute stroke: A safe and hu-
However, in our study only 16 patients had brainstem mane screening tool. Age Ageing 2000;29:495-499.
18. Ramsey D, Smithard DG, Kalra L. Early assessment of
lesions and we did not identify precise lesion location. dysphagia and aspiration risk in acute stroke patients.
This could have reduced the statistical significance on Stroke 2003;34:1252-1257.
this subgroup. 19. Smith CJ, Emsley HC, Libetta CM, et al. The Oxfordshire
Thus, whether lesion size, location, or type of stroke in community stroke project classification in the early hours
conjunction with clinical or neurocognitive defects are re- of ischemic stroke and relation to infarct site and size on
cranial computed tomography. J Stroke Cerebrovasc Dis
lated to dysphagia and aspiration remains debated and 2001;10:205-209.
further study is needed to clarify this question. 20. Bamford J, Sandercock P, Dennis M, et al. Classification
and natural history of clinically identifiable subtypes of
cerebral infarction. Lancet 1991;337:1521-1526.
References 21. Schroeder MF, Daniels SK, McClain M, et al. Clinical and
cognitive predictors of swallowing recovery in stroke.
1. Barer DH. The natural history and functional conse- J Rehabil Res Dev 2006;43:301-310.
quences of dysphagia after hemispheric stroke. J Neurol 22. Finestone HM, Greene-Finestone LS, Wilson ES, et al.
Neurosurg Psychiatry 1989;52:236-241. Malnutrition in stroke patients on the rehabilitation
DYSPHAGIA FOLLOWING STROKE 335

service and at the follow-up: Prevalence and predictors. 32. Smithard DG, O’Neill PA, Martin DF, et al. Aspiration fol-
Arch Phys Med Rehabil 1995;76:310-316. lowing stroke: Is it related to the side of the stroke? Clin
23. Daniels SK, Schroeder MF, McClain M, et al. Dysphagia Rehabil 1997;11:73-76.
after stroke: Development of a standard method to exam- 33. Daniels SK, Foundas AL, Iglesia GC, et al. Lesion site in
ine swallowing recovery. J Rehabil Res Dev 2006; unilateral stroke patients with dysphagia. J Stroke Cere-
43:347-356. brovasc Dis 1996;6:30-34.
24. Gates J, Hartnell GG, Gramigna GD. Videofluoroscopy 34. Alberts MJ, Horner J, Gray L, et al. Aspiration after
stroke: Lesion analysis by brain MRI. Dysphagia 1992;
and swallowing studies for neurologic disease: A primer.
7:170-173.
Radiographics 2005;26:e22.
35. Langdon PC, Lee AH, Binns CW. Dysphagia in acute is-
25. Rosenbek JC, Robbins JA, Roecker EB, et al. A penetra- chemic stroke: Severity, recovery and relationship to
tion-aspiration scale. Dysphagia 1996;11:93-98. stroke subtype. J Clin Neurosci 2007;14:630-634.
26. Gottlieb D, Kipnis M, Sister E, et al. Validation of the 50 36. Broadley S, Croser D, Cottrel J, et al. Predictors of pro-
mL 3 drinking test for evaluation of post-stroke dyspha- longed dysphagia following acute stroke. J Clin Neurosci
gia. Disabil Rehabil 1996;18:529-532. 2003;10:300-305.
27. Ickenstein GW, Stein J, Ambrosi D, et al. Predictors of sur- 37. Paciaroni M, Mazzotta G, Corea F, et al. Dysphagia fol-
vival after severe dysphagic stroke. J Neurol 2005; lowing stroke. Eur Neurol 2004;51:162-167.
252:1510-1566. 38. Barros AF, Fabio SR, Furkim AM. Relation between clin-
28. Iizuka M, Reding M. Use of percutaneous endoscopic ical evaluation of deglutition and computed tomography
gastrostomy feeding tubes and functional recovery in in acute ischemic stroke patients. Arq Neuropsiquiatr
stroke rehabilitation: A case-matched controlled study. 2006;64:1009-1014.
Arch Phys Med Rehabil 2005;86:1049-1052. 39. Kim H, Chung CS, Lee KH, et al. Aspirations subsequent
29. DePippo KL, Holas MA, Reding MJ. The burke dyspha- to a pure medullary infarction: Lesion sites, clinical vari-
ables and outcome. Arch Neurol 2000;57:478-483.
gia screening test: Validation of its use in patients with
40. Daniels SK, Foundas AL. Lesion localization in acute
stroke. Arch Phys Med Rehabil 1996;75:1284-1286.
stroke patients with risk of aspiration. J Neuroimaging
30. Chua KSG, Kong KH. Functional outcome in brainstem
1999;9:91-98.
stroke patients after rehabilitation. Arch Phys Med Reha- 41. Lawrence ES, Coshall C, Dundas R, et al. Estimates of the
bil 1996;77:194-197. prevalence of acute stroke impairments and disability in
31. Goldstein LB, Jones MR, Matchar DB, et al. Improving the a multiethnic population. Stroke 2001;32:1279-1284.
reliability of stroke subgroup classification using the trial 42. Terao S, Miura N, Osano Y, et al. Multiple cerebellar in-
of ORG 10172 in acute stroke treatment (TOAST) criteria. farcts: Clinical and pathophysiologic features. J Stroke
Stroke 2001;32:1091-1097. Cerebrovasc Dis 2005;14:193-198.