Dysphagia and Aspiration Post Stroke

Document Sample
Dysphagia and Aspiration Post Stroke Powered By Docstoc
					15. Dysphagia and Aspiration Post Stroke
Robert Teasell MD, Norine Foley MSc, Rosemary Martino, PhD, Sanjit Bhogal MSc, Mark Speechley PhD

Key Points
There is a high incidence of dysphagia and aspiration following acute The Evidence-Based
stroke.                                                               Review of Stroke
                                                                               Rehabilitation (EBRSR)
                                                                               reviews current
VMBS studies are the only sure way of diagnosing dysphagia and                 practices in stroke
aspiration.                                                                    rehabilitation.

The incidence of silent aspiration following acute stroke is high.             Contacts:
                                                                               Dr. Robert Teasell
The risk of developing pneumonia following stroke is proportional to           801 Commissioners
                                                                               Road East
the severity of aspiration.
                                                                               London, Ontario,
All stroke survivors should remain NPO until a trained assessor has
                                                                               N6C 5J1
assessed swallowing ability.
Following a failed screening, a referral to a Speech-Language
Pathologist should be made for further assessment and management.
Feeding assistance should be provided by an individual trained in low-
risk feeding strategies. Individuals with dysphagia should feed
themselves whenever possible.

Dysphagia diets, consisting of texture-modified solid foods and
partially thickened fluids may help to reduce the incidence of
aspiration pneumonia.

Treatments with Nifedipine, transcranial magnetic stimulation and
head rotation techniques can be used to improve swallowing
mechanics, while thermal stimulation may not.

Enteral tube feeding may be necessary when stroke patients fail to
meet their nutritional needs orally. There is no difference in the
outcomes of death or poor outcome associated with the use of either
nasogastric or gastro-enteric feeding tubes.

It is uncertain if the use of electrical stimulation improves swallowing
function post stroke.

Last updated August 2010

 15.   Dysphagia                                                                     pg. 1 of 60
Table of Contents

Key Points .......................................................................................... 1
Table of Contents ................................................................................. 2
15. Dysphagia and Aspiration Post Stroke.................................................. 3
  15.1 Normal Swallowing .................................................................................3
  15.2 Pathophysiology of Dysphagia..................................................................4
  15.3 Aspiration Associated with Dysphagia ........................................................4
    15.3.1 Silent Aspiration Post Stroke ................................................................. 5
  15.4 Incidence of Dysphagia Post Stroke ...........................................................5
    15.4.1 Acute Phase of Stroke......................................................................... 5
    15.4.2 Incidence of Aspiration Following Stroke ................................................... 7
    15.4.3 The Prevalence of Dysphagia in the Rehabilitation Stage Post Stroke ............... 9
  15.5 Pneumonia and Aspiration Post Stroke .................................................... 10
    15.5.1 Defining Aspiration Pneumonia ............................................................ 11
    15.5.2 Relationship Between Pneumonia and Dysphagia/Aspiration ........................ 12
  15.6 Non-Instrumental Methods for Screening and Assessment of Dysphagia Following
  Stroke ...................................................................................................... 14
    15.6.1 Clinical Screening Methods................................................................. 14
    15.6.2 The Water Swallowing Test ................................................................ 17
    15.6.3 Swallowing Provocation Test (SPT) ....................................................... 19
    15.6.4 The Bedside Clinical Examination for Assessment of Dysphagia..................... 20
    15.6.5 Other Methods................................................................................ 21
  15.7 Instrumental Methods Used in the Detection of Dysphagia/Aspiration.............. 22
    15.7.1 VMBS Examination .......................................................................... 22
    15.7.2 Flexible Endoscopic Evaluation of Swallowing (FEES)................................. 23
    15.7.3 Pulse Oximetry ............................................................................... 25
  15.8 Management of Aspiration Post Stroke ..................................................... 26
    15.8.1 Management Strategies for Dysphagia ................................................... 27
    15.8.2 Best Practice Guidelines for Managing Dysphagia...................................... 27
    15.8.3 Dysphagia Screening Protocols............................................................ 29
    15.8.4 Low-Risk Feeding Strategies for Dysphagia ............................................ 29
  15.9 Specific Interventions to Manage Dysphagia .............................................. 31
    15.9.1 Dietary Modifications........................................................................ 31
    15.9.2 Swallowing Treatment Programs ......................................................... 34
    15.9.3 Non-Oral Feedings........................................................................... 37
    15.9.4 Selection of Feeding Tubes ................................................................ 39
    15.9.5 Alternative Interventions .................................................................... 43
  15.10 Summary .......................................................................................... 51
  References................................................................................................ 54

15.   Dysphagia                                                                                      pg. 2 of 60
                                              in stroke survivors was associated
15. Dysphagia and Aspiration                  with the length of hospitalization.
                                              Detection of aspiration, both silent and
Post Stroke                                   audible, and subsequent adaptive
Dysphagia is defined as difficulty with       management strategies are regarded
swallowing and is a common                    as important in the prevention of
complication of stroke. The incidence         pneumonia (Horner and Massey
rates are reported to be between 29-          1988a, Horner et al. 1988b,
67% in acute stroke patients (Martino         Logemann 1983, Teasell et al. 1996,
et al. 2005). Some of the variability is      Tobin 1986, Veis and Logemann
related to differences in the timing          1985). Management of dysphagia
and method of swallowing                      largely focuses on strategies to avoid
assessment. The presence of                   aspiration following stroke.
dysphagia can be identified on the
basis of clinical or radiographic             15.1 Normal Swallowing
examinations, or both.
                                              Swallowing has four sequential
The presence of dysphagia in stroke           coordinated phases: the oral
survivors has been associated with            preparatory phase, the oral propulsive
increased mortality and morbidities           phase, the pharyngeal phase and the
such as malnutrition, dehydration and         esophageal phase. Each of the phases
pulmonary compromise (Smithard et             of a normal swallow is described below
al. 1996, Barer 1989, Kidd et al. 1995,       (Jean et al. 2001).
Finestone et al. 1995, Teasell et al.
1994, Gordon et al. 1987, Schmidt et          Oral Preparatory Phase. During this
al. 1994, Sharma et al. 2001).                phase, food in the oral cavity is
Evidence indicates that detecting and         manipulated and masticated in
managing dysphagia in acute stroke            preparation for swallowing. The back
survivors improves outcomes such as           of the tongue controls the position of
reduced risk of pneumonia, length of          the food, preventing it from falling
hospital stay and overall healthcare          into the pharynx.
expenditures (Smithard et al. 1996).
                                              Oral Propulsive Phase. During the
Aspiration following stroke, the most         oral propulsive, the tongue transfers
clinically significant symptom of             the bolus of food to the pharynx,
dysphagia, has long been associated           triggering the pharyngeal swallow.
with pneumonia, sepsis and death.
Silver et al. (1984) and Bounds et al.        Pharyngeal Phase. During the
(1981) reported that pneumonia was            pharyngeal phase, complex and
the second most common cause of               coordinated movements of the tongue
death during the acute phase of a             and pharyngeal structures propel the
stroke, with up to 20% of individuals         bolus from the pharynx into the
with stroke-related dysphagia dying           esophagus. The closing of the vocal
during the first year post stroke from        cords and the backward movement of
aspiration pneumonia (Schmidt et al.          the epiglottis prevents food or liquid
1988). Steele (2002) found that the           from entering the trachea.
number of swallowing difficulties seen
                                              Esophageal Phase. During the
                                              esophageal phase of swallowing,

15.   Dysphagia                                                             pg. 3 of 60
coordinated contractions of the                       of dysphagia include: Choking on
esophageal muscle move the bolus                      food, coughing during meals, drooling
through the esophagus towards the                     or loss of food from mouth, pocketing
stomach.                                              on food in cheeks, slow, effortful
                                                      eating, difficulty swallowing pills,
15.2 Pathophysiology of Dysphagia                     avoiding food or fluids, complaining of
                                                      food sticking in throat, problems
Dysphagia post stroke has long been                   swallowing, reflux or heartburn
attributed to pharyngeal muscular                     (Schmidt et al. 1994). Table 15.1
dysfunction and incoordination,                       summarizes the results of studies
secondary to central nervous system                   assessing the pathophysiology.
loss of control. Signs and symptoms

Table 15.1 Pathophysiology of Dysphagia Post Stroke
   Author                   Methods                                       Results
Veis and     38 stroke patients consecutively         50% of patients demonstrated reduced lingual
Logemann     referred for VMBS examination for        control, 82% a delayed reflex, 58% reduced
1985         suspicion of swallowing disorders        pharyngeal peristalsis, 5% reduced laryngeal
USA          within 4 months of stroke. VMBS          adduction, 5% crichopharyngeal dysfunction.
No Score     studies were used to assess oral and     76% of patients demonstrated more than one
             pharyngeal functioning and to identify   swallowing disorder. 32% of patients aspirated.
             motility disorders. 3 consistencies
             were tested: liquid, paste and cookie.
Robbins et   The swallowing patterns of 20 first -    Patients with left hemisphere strokes had
al. 1993     ever MCA stroke patients were            longer pharyngeal transit duration times
USA          compared with 40 control subjects.       compared to controls. Patients with right
No Score                                              hemisphere strokes demonstrated longer
                                                      pharyngeal stage durations and higher
                                                      incidences of laryngeal penetration and
                                                      aspiration of liquid. Anterior lesion subjects
                                                      demonstrated significantly longer swallowing
                                                      durations on most variables compared to both
                                                      normal and posterior lesion subjects.

 Conclusions Regarding the
 Pathophysiology of Dysphagia                         15.3 Aspiration Associated with
 Dysphagia post stroke is characterized               Dysphagia
 by a delay and reduced function in the               Aspiration is defined as "entry of
 pharyngeal phase of swallowing.                      material into the airway below the
 Although the incidence of dysphagia is               level of the true vocal cords". Since
 more common following brainstem or                   many stroke patients with dysphagia
 bilateral hemispheric stroke, it                     do not aspirate, the two terms are not
 frequently occurs following unilateral
                                                      synonymous, although they are closely
 hemispheric strokes.
                                                      associated. The diagnosis of aspiration
                                                      should be suspected when the stroke
  Dysphagia is characterized by reduced               patient has any of the following: a
   coordination of pharyngeal muscles.                subjective complaint of trouble

15.   Dysphagia                                                                           pg. 4 of 60
swallowing, an abnormal chest x-ray,                  minimal amount of fluid, a small cup of
congested voice quality, or a delay in                water is carefully introduced. The full
voluntary initiation of the swallow                   assessment is described elsewhere
reflex and coughing during or after                   (Smithard et al. 1996). While all
swallowing (Horner et al 1988b).                      stroke patients are potential
Diagnosis is initially established                    aspirators, there are certain
through clinical assessment involving                 identifiable risk factors that have been
an oral motor examination followed by                 recognized as greatly increasing the
the introduction of one or several                    likelihood of aspiration. These clinical
teaspoons of water. If patients are                   risk factors are listed in Table 15.2.
able to successfully swallow this

Table 15.2 Risk Factors for Aspiration Post-Stroke

 •      Brainstem Stroke
 •      Difficulty swallowing oral secretion
 •      Coughing/throat clearing or wet, gurgly voice quality after swallowing water
 •      Choking more than once while drinking 50 ml of water
 •      Weak voice and cough
 •      Wet-hoarse voice quality
 •      Recurrent lower respiratory infections
 •      Low-grade fever or leukocytosis
 •      Auscultatory evidence of lower lobe congestion
 •      Immunocompromised state

15.3.1 Silent Aspiration Post Stroke                  developing complications. Since the
In addition to overt signs of aspiration,             condition was not diagnosed,
such as chocking or coughing, a                       precautions to decrease aspiration risk
substantial number of patients                        would often not be employed. Silent
experience silent aspiration,                         aspiration should be suspected in the
highlighting the utility of using VMBS                stroke patient with recurrent lower
studies. "Silent aspiration" is defined               respiratory infections, chronic
as "penetration of food below the level               congestion, low- grade fever or
of the true vocal cords, without cough                leukocytosis (Muller-Lissner et al.
or any outward sign of difficulty"                    1982). Clinical markers of silent
(Linden and Siebens 1983). Detailed                   aspiration may include a weak voice or
clinical swallowing assessments were                  cough or a wet-hoarse quality after
shown to under-diagnose or miss these                 swallowing.
cases of aspiration (Horner and Massey
1988a, Horner et al. 1988b, Splaingard
                                                      15.4 Incidence of Dysphagia Post
et al. 1988, Terre & Mearin 2006). In                 Stroke
particular, the presence or absence of                15.4.1 Acute Phase of Stroke
a gag reflex failed to distinguish
                                                      Table 15.3 presents results from a
aspirating from non-aspirating stroke
                                                      variety of studies that used clinical
patients (Horner and Massey 1988a,
                                                      methods to assess swallowing among
Horner et al. 1988b, Splaingard et al.
                                                      acute and rehabilitating stroke
1988). Silent aspirators were
considered to be at increased risk of

15.   Dysphagia                                                                        pg. 5 of 60
Table 15.3 Incidence of Dysphagia Post Stroke (Acute)
   Author                           Methods                                         Results
Gordon et al.    91 consecutive stroke patients were evaluated 41 (45%) of the patients had
1987             with a standardized swallowing test for the   evidence of dysphagia.
UK               presence of dysphagia. 82/90 (92%) were
No Score         evaluated within 4 days of stroke onset.

Wade et al.      452 consecutive, conscious acute stroke            194 (43%) of patients were
1987             patients were evaluated within 7 days of onset     considered dysphagic.
UK               of symptoms. Their ability to swallow water
No Score         from a cup was evaluated.
Barer 1989       357 stroke patients selected from a stroke         105 (29%) patients initially presented
UK               registry participating in the “BEST” study with    with dysphagia. By 1 month, 6/277
No Score         onset of symptoms within 48 hrs, single-           (2%) of survivors were still dysphagic.
                 hemisphere involvement, able to take oral          At 6 months, 1/248 (0.4%) of those
                 medications, no pre-stoke impairments and no       assessed remained dysphagic.
                 cardiac conditions were followed. Ability to
                 swallow 10 mL of water from a cup was
Odderson et      124 consecutive, ischemic stroke patients          48 (39%) of patients failed the initial
al. 1995         were assessed within 24 hours using a clinical     swallow screen. 21 of patients with
USA              swallowing screen, assessing voice quality,        dysphagia recovered their swallowing
No Score         ability to handle oral secretions and ability to   function by discharge.
                 swallow ice chips or water briskly.
Nilsson et al.   100 consecutive, acute stroke patients were        Only 72 patients were able to reliably
1998             assessed for dysphagia within 24 hours of          respond to questions. 14 patients
Sweden           admission. Patients were questioned about          (19%) complained of dysphagia.
No Score         swallowing complaints and examined with a
                 repetitive oral suction swallow test (ROSS).
Daniels et al.   55 stroke patients consecutively admitted to a     Dysphagia was present in 36 (65%) of
1998             VA medical centre. All patients received a         patients. Aspiration occurred in 21
USA              bedside and VMBS evaluation within 5 days of       (38%) patients.
No Score         admission.
Mann et al.      The swallowing function of 128 hospital-           Using VMBS a median of 10 days
1999             referred patients with acute stroke was            following stroke, 82 (64%) of patients
Australia        evaluated clinically and with VMBS studies.        were diagnosed with dysphagia, 28
No Score         Patients were followed for 6 months.               (22%) aspirated. Using a clinical
                                                                    exam administered a median of 3
                                                                    days following stroke, the incidence of
                                                                    dysphagia and aspiration were 51%
                                                                    and 50%, respectively.
Gosney et al.    203 patients with first ever stroke were           On admission 58 (29%) were
2006             recruited from the acute stroke assessment         considered to be dysphagic.
UK               units from 3 hospitals within 24 hours of onset
6 (RCT)          of symptoms. Patients were evaluated by
                 either a bedside evaluation or a water-
                 swallowing test.
Smithard et      A population-based long-term follow-up of       567 (44%) patients were determined
al. 2007         1,288 persons with first time stroke.           to be dysphagic at the point of first

15.   Dysphagia                                                                                 pg. 6 of 60
UK               Dysphagia was assessed by clinical exam           assessment.
No Score         within 1 week of stroke. Patients were
                 followed up at 3 months and yearly for 5

The studies reviewed above assessed                         sample studied and the assessment
swallowing status in the acute phase                        method used.
of stroke assessed using both clinical
and VMBS examination. Among these                            There is a high incidence of
studies, the incidence of dysphagia                          dysphagia following acute stroke.
ranged from 19% to 65%.
                                                          15.4.2 Incidence of Aspiration
 Conclusions Regarding the Incidence                      Following Stroke
 of Dysphagia (Acute)
                                                          Several studies have estimated the
 The incidence of dysphagia appears to                    incidence of aspiration and silent
 be quite high following acute stroke
                                                          aspiration post stroke using a
 with between one third and two-thirds
                                                          combination of clinical and
 of patients affected, depending on the
                                                          radiographic techniques.
                                                          (see Table 15.4)

Table 15.4 Incidence of Aspiration and Silent Aspiration Post Stroke
   Author                     Methods                                        Results
Veis and        38 stroke patients consecutively        50% of patients demonstrated reduced lingual
Logemann        referred for VMBS studies for           control, 82%, a delayed reflex, 58% reduced
1985            suspicion of swallowing disorders       pharyngeal peristalsis, 5% reduced laryngeal
USA             within 4 months of stroke. VMBS         adduction, 5% crichopharyngeal dysfunction.
No Score        studies to assess oral and              76% of patients demonstrated more than one
                pharyngeal functioning and to           swallowing disorder. 32% of patients aspirated.
                identify motility disorders. 3
                consistencies were tested: liquid,
                paste and cookie.
Horner et al.   47 stroke patients referred for    51% of patients aspirated on at least one
1988 (b)        swallowing evaluation on suspicion consistency. 54% of the aspirators were silent
USA             of dysphagia. To evaluate the      aspirators. Aspiration was not limited to
No Score        clinical correlates of dysphagia,  brainstem or bilateral lesions. Medical-clinical
                patients received both a clinical and
                                                   abnormalities appeared to be more frequent in
                videofluoroscopic evaluation of    patients who aspirated although no statistical
                swallowing function. Liquid, paste analysis was performed. The presence of a
                and cookie consistencies were      delayed swallow reflex and reduced peristalsis
                tested. 33 patients were tested    frequently resulted in aspiration. Poor oral
                within the first month post stroke.motility did not.
Chen et al.     46 consecutive patients with clinical
                                                   Dysphagia was confirmed by VMBS examination
1990            symptoms of dysphagia within one-  in all cases. Mild swallowing impairment was
USA             month of stroke were referred for  identified in 18 (39%) patients, moderate
No Score        VMBS examination.                  dysfunction in 23 (50%) and severe problems in
                                                   5 (11%) patients. There were 24 episodes of
Splaingard et 107 patients referred for evaluation 40% of patients aspirated during VMBS study.

15.   Dysphagia                                                                               pg. 7 of 60
al.           for possible swallowing dysfunction   Bedside evaluations identified only 42% of proven
1988          from a general rehabilitation ward,   aspirators. Silent aspiration, not detected on
USA           including 87 stroke patients. The     bedside evaluation was noted in 20% of patients.
No Score      results of a bedside swallowing       Bedside assessments identified 58/64 (90%) of
              evaluation were compared with         non-aspirators.
              VMBS results by blinded evaluators.
Kidd et al.   60 consecutive stroke patients        42% of patients aspirated on initial VMBS.
1995          admitted to a teaching hospital.      42% of patients were unable to complete the
UK            Patients received a water-            water-swallowing test. Of these, 80% were
No Score      swallowing test and VMBS study        aspirators. 32% of patients developed a
              within 72 hrs of stroke onset and     respiratory tract infection (RTI) within 14 days.
              were re-evaluated at 3 months.        89% of RTIs occurred in aspirating patients. 42
                                                    patients were re-examined at 3 months. 14% of
                                                    patients continued to experience impaired
                                                    pharyngeal sensation. An abnormal water-
                                                    swallowing test was reported in 7% of the
                                                    remaining patients. 8% of patients initially
                                                    presenting with a positive VMBS result also had a
                                                    positive follow-up test. These same patients
                                                    developed a respiratory tract infection between
                                                    days 14 and 90. 5 patients were silent aspirators,
                                                    accounting for 20% of all aspirators.
Smithard et 121 stroke patients consecutively       50% of the patients were considered to have an
al. 1996       admitted to an urban hospital.       unsafe swallow based on bedside evaluation
UK             Patients received both bedside and alone. 94 patients had a VMBS study. Of these,
No Score       VMBS evaluations within 3 days of    20 (16.5%) patients aspirated. Increased
               stroke, when feasible.               mortality, lower Barthel scores and increased
                                                    frequency of discharge to institutionalized care at
                                                    6 months were reported more often in patients
                                                    with an unsafe swallow. However, these
                                                    outcomes were not associated with a positive
                                                    VMBS study result. 22 patients did not receive a
                                                    6-month follow-up.
Daniels et al. 59 consecutively admitted ischemic 44/59 patients (74.6%) were dysphagic based on
1997           stroke patients received a clinical  VMBS results. Dysphonia, dysarthria, abnormal
USA            and VMBS swallowing evaluation       volitional cough and cough after swallow were all
No Score       within 5 days of admission.          significantly predictive of dysphagia severity.
Daniels et al. 55 stroke patients consecutively     Dysphagia was present in 65% of patients.
1998           admitted to a VA medical centre. All Aspiration occurred in 21 (38%) patients. Of
USA            patients received a bedside and      these, 14 aspirated silently (67% of aspirators).
No Score       VMBS evaluation within 5 days of     Dysphagia was present in 65% of patients.
               admission.                           Both abnormal volitional coughing and cough with
                                                    swallow were highly predictive of aspiration. One
                                                    patient developed aspiration pneumonia during
Mann et al.    The swallowing function of 128       Using VMBS a median of 10 days following stroke,
1999           hospital-referred patients with      82 (64%) of patients were diagnosed with
Australia      acute stroke was evaluated           dysphagia, 28 (22%) aspirated. Using a clinical
No Score       clinically and with VMBS studies.    exam administered a median of 3 days following
               Patients were followed for 6         stroke, the incidence of dysphagia and aspiration
               months.                              were 51% and 50%, respectively.
Kim et al.     23 patients with isolated medullary Ten (44%) of the 23 patients manifested
2000           infarctions were assessed using      aspiration on swallowing. Using criteria including

15.   Dysphagia                                                                            pg. 8 of 60
USA               VMBS studies within two weeks of    dysphonia, soft palate dysfunction, and facial
No Score          stroke. From the results of the     hypesthesia were used to discriminate between
                  VMBS studies, 2 patient groups      those with and without aspiration with 95.7%
                  were formed: one with aspiration    accuracy.
                  and the other without aspiration.
                  The clinical variables related to
                  aspiration and outcome measures
                  were also explored.

Discussion                                                 only reliably detectable through VMBS
                                                           studies. Between one-third and one-
The incidence of aspiration identified                     half of aspirators are “silent”
using VMBS studies, assessed during                        aspirators.
the acute stage of stroke, ranged from
30% to 51%. The incidence of silent                         Aspiration following stroke is very
aspiration was reported in five studies.                    common.
The incidence of silent aspirators                          The incidence of silent aspiration
ranged from 8% (Kidd et al. 1995) to                        following acute stroke is high.
27% (Horner et al. 1988b).

 Conclusions Regarding the Incidence
 of Aspiration Using VMBS                                15.4.3 The Prevalence of Dysphagia
                                                         in the Rehabilitation Stage Post
 The incidence of aspiration in the acute                Stroke
 phase of stroke varies from 30% to
 51%.                                                    Few studies have examined the
                                                         prevalence of dysphagia that may
 Conclusions Regarding Silent
                                                         persist past the acute stage of stroke.
                                                         The results from 5 studies that were
 9 to 27% of acute stroke patients are                   identified, specific to the rehabilitation
 silent aspirators, a condition which is                 period are presented in Table 15.5.

Table 15.5 Prevalence of Dysphagia at Admission to Rehabilitation Post Stroke
   Author                            Methods                                       Results
DePippo et al.    139 consecutive patients admitted to an           82 (59%) of patients failed the
1994              inpatient rehabilitation unit a mean of 5 weeks   screening tests.
USA               following stroke. Patients were evaluated
No Score          using the Burke Dysphagia Screening test (a
                  water swallowing test).
Gottlieb et al.   180 consecutive rehab patients assessed an        Dysphagia was diagnosed in 28% of
1996              average of 14 days post stroke using a            patients.
Israel            bedside technique, which included a water
No Score          swallowing test (50 mL). A cough during
                  drinking was considered positive.
Terre &           138 consecutive patients admitted to a            Dysphagia was clinically suspected in
Mearin 2006       rehabilitation hospital recovering from a         64 (46%) of patients. Clinical
Spain             severe, first-ever strokes were evaluated         examination showed that 44% had
No Score          clinically and through videofluoroscopy.          impaired gag reflex, 47% coughed

15.   Dysphagia                                                                                pg. 9 of 60
                  Evaluations were conducted a mean of 3           during oral feeding, and 13%
                  months following stroke                          demonstrated changes in voice after
                                                                   swallowing. 42 (30%) patients
                                                                   demonstrated pharyngeal aspiration
                                                                   on VMBS. Of these, 21 (50%) were
                                                                   episodes of silent aspiration.
Poels et al.      69 stroke patients without aphasia admitted to Eating difficulties (including
2006              stroke rehabilitation an average of 34 days      swallowing difficulties) were present
The               following acute stroke. Dysphagia was            in 30 (43%) of patients.
Netherlands       assessed using structured observations of
No Score          eating difficulties.
Falsetti et al.   151 consecutively-admitted patients admitted 62 (41%) of patients were dysphagic,
2009              to a neurorehabilitation unit an average of 14 based on the results from the clinical
Italy             days following stroke received a 3-step clinical exam. 49/151 patients, 79% of whom
No Score          exam, which included 2 water swallowing          were identified as dysphagic based on
                  components (bolus of differing amounts)          clinical exam, received a VFS study.
                  within one day of admission and a VFS exam       Aspiration and silent aspiration were
                  for those who failed any portion of the          detected in 21 (43%) and 13 (26.5%)
                  screening test.                                  patients, respectively.

Discussion                                               (Chua et al. 1996- 40%, Teasell et al.
                                                         2002-55%, Meng et al. 2000-81%).
Among five samples of unselected
patients entering stroke rehabilitation,                   Conclusions Regarding the Prevalence
the prevalence of dysphagia ranged                         of Dysphagia in the Rehabilitation
from 28-59%, using different                               Stage of Stroke
assessment techniques. If studies
which limit patient selection to those                     A high percentage of patients enter
recovering from brainstem stroke, the                      rehabilitation with persistent
percentages are significantly higher                       dysphagia.

                                                         the immune state or general health of
15.5 Pneumonia and Aspiration                            the stroke patient. Sellars et al.
                                                         (2007) prospectively evaluated 412
Post Stroke                                              stroke patients for up to 3 months
Those patients who aspirate over 10%                     following stroke. Over this period,
of the test bolus or who have severe                     there were 160 cases of either
oral and/or pharyngeal motility                          confirmed or suspected pneumonias.
problems on VMBS studies are                             Independent predictors of pneumonia
considered at high risk for pneumonia                    were age >65 years, dysarthria or no
(Logemann 1983, Milazzo et al. 1989).                    speech due to aphasia, a modified
In many cases, it is difficult to                        Rankin Scale score ≥4, an Abbreviated
practically assess whether 10% or                        Mental Test score <8, and failure on
more of the test bolus has been                          the water swallow test. The presence
aspirated. Nevertheless, the degree                      of 2 or more of these risk factors
of aspiration seen on VMBS study is a                    carried 90.9% sensitivity and 75.6%
critical determinant of patient                          specificity for the development of
management. Predicting whether a                         pneumonia.
patient will develop pneumonia post
aspiration is, to some extent,                           The importance of the diagnosis and
dependent on other factors such as                       management of aspiration post stroke

15.   Dysphagia                                                                            pg. 10 of 60
has been driven by the purportedly                      in almost half of normal subjects
causal relationship between aspiration                  (Finegold 1991, Huxley et al. 1978).
and pneumonia (Brown and                                Aspiration pneumonia is thought to
Classenberg 1973, Hanning et al.                        occur when the lung's natural
1989, Holas et al. 1994, Johnson et al.                 defences are overwhelmed when
1993). In turn, mortality following a                   excessive and/or toxic gastric contents
stroke as a consequence of pneumonia                    are aspirated, leading to a localized
(presumably due to aspiration) has                      infection or a chemical pneumonitis.
been reported as high as 3% within                      Factors associated with an increased
the first 3 months (Kidd et al. 1995)                   risk of aspiration pneumonia include:
and 6% within the first year (Hanning                   dysphagia related factors due to
et al. 1989). Aspiration pneumonia                      stroke (see Table 15.6), as well as
has therefore been regarded as                          reduced levels of consciousness, a
important because of its significant                    tracheostomy, gastric reflux or
contribution to morbidity and mortality                 emesis, nasogastric tubes (due to
(Arms et al. 1974, Gordon et al. 1987,                  mechanical interference with the
Hanning et al. 1989, Johnson et al.                     cardiac sphincter), and a compromised
1993, Logemann 1983, Silver et al.                      immune system (Finegold 1991).
1984, Veis and Logemann 1985).                          However, it remains uncertain to what
                                                        degree the aspiration of colonized
Aspiration alone is not sufficient to                   oropharyngeal contents contributes to
cause pneumonia. Aspiration of small                    pneumonia (Langdon et al. 2009).
amounts of saliva occurs during sleep

Table 15.6 Factors More Likely to be Associated with Aspiration Pneumonia
Following Stroke
 •      Brainstem stroke
 •      Aspiration on VMBS (risk greater if aspirates over 10% of barium laced test material)
 •      Aspiration of thick fluids or solids
 •      Slower pharyngeal transit time on VMBS

15.5.1 Defining Aspiration Pneumonia                    pneumonia influences its incidence.
                                                        Much of the variability in incidence of
Clinical criteria for aspiration                        aspiration among studies can be
pneumonia across studies have proven                    accounted for by differences in the
to be variable (Table 15.7). Obviously                  inclusion criteria for the diagnosis of
the criteria used for defining                          pneumonia.

Table 15.7 Criteria For Defining Pneumonia in Stroke
    Author/ Year                                                Criteria
Johnson et al. 1993    Aspiration pneumonia was defined by either segmental consolidation or infiltrate on
USA                    chest x-ray or clinical diagnosis which included an episode of respiratory difficulty with
No Score               segmental moist rales on auscultation and two other symptoms including temp
                       >100 °F, WBC >10,000 or hypoxia.
DePippo et al. 1994    Pneumonia was diagnosed by a positive chest x-ray or the presence of at least three
USA                    of the following: temp > 100 °F, drop in PO2 > 10 torr, presence of WBC in sputum
No Score               and/or positive sputum culture for pathogen.

15.   Dysphagia                                                                              pg. 11 of 60
Holas et al. 1994     Pneumonia was diagnosed by a positive chest x-ray or the presence of at least three
USA                   of the following: temp > 100 °F, drop in PO2 > 10 torr, presence of WBC in sputum
No Score              and/or positive sputum culture for pathogen.
Kidd et al. 1995      Diagnosis of pneumonia was based on the production of sputum in conjunction with
UK                    the development of crackles on auscultation, with or without the presence of fever or
No Score              leucocytosis.
Smithard et al.       Chest infection was diagnosed on the presence of at least two of the following:
1996                  tachypnea (> 22/min), tachycardia, aspiratory crackles, bronchial breathing or
UK                    antibiotic usage.
No Score
Teasell et al. 1996   The criteria for pneumonia included radiological evidence of consolidation, and at least
Canada                one other clinical feature including granulocytosis, temp >38°C and/or shortness of
No Score              breath.
Dziewas et al. 2004   Pneumonia was diagnosed on the basis of 3 of the following indicators: temp >38°C,
Germany               productive cough with purulent sputum, abnormal respiratory exam including
No Score              tachypnea, (> 22 breaths/min), tachycardia, inspiratory crackles, bronchial
                      breathing, abnormal chest x-ray, arterial hypoxemia (PO2 < 9.3 kPa) and a positive
                      gram stain.
Carnaby et al. 2006   Pneumonia was diagnosed on the basis of 3 of the following indicators: temp >38°C,
USA                   productive cough, abnormal respiratory exam including tachypnea, (> 22
8 (RCT)               breaths/min), tachycardia, inspiratory crackles, bronchial breathing, abnormal chest
                      x-ray, arterial hypoxemia (PO2 < 9.3 kPa), culture of a relevant pathogen; positive
                      chest radiography.
                                                       swallowing response, assessed by
15.5.2 Relationship Between                            EMG activity and direct observation
Pneumonia and Dysphagia/Aspiration                     was greater than 20 sec. In contrast,
                                                       the latency of response was less than
A relationship between pneumonia and                   4 seconds among patients without
dysphagia/aspiration has been                          dysphagia. The association between
reasonably well established despite                    pneumonia and both dysphagia and
variability among studies. Nakajoh et                  aspiration is examined among a series
al. (2004) have suggested that                         of studies using odds ratios. The
attenuated cough reflexes also                         results are presented in tables 15.8
increases a patients’ risk of                          And 15.9 and graphically in figures
pneumonia. The incidence of                            15.1 and 15.2. In all cases the
pneumonia among dysphagic,                             incidence of pneumonia was higher
bedridden patients who had suffered                    among patients with dysphagia and/or
from a stroke for at least 6 months                    aspiration
was 9/14 (63%). The latency of the

15.   Dysphagia                                                                           pg. 12 of 60
Table 15.8 Relationship Between Dysphagia and Pneumonia
        Study          Incidence of Pneumonia Among Patients with   OR (95% CI, fixed effects model)
                                  and without Dysphagia
Gordon et al. 1987                    7/37 vs. 4/50                      2.63 (0.72 to 9.96)
De Pippo et al. 1994                 10/82 vs. 1/57                      7.78 (0.97 to 62.6)
Gottlieb et al. 1996                 9/50 vs. 9/130                      2.95 (1.10 to 7.94)
Smithard et al. 1996                 20/60 vs. 9/57                      2.67 (1.09 to 6.50)
Reynolds et al. 1998                 18/69 vs. 3/33                     3.53 (0.96 to 12.99)
Teasell et al. 2002                    5/11 vs. 0/9                               -
Falsetti et al. 2009                  1/89 vs. 8/62                     13.04 (1.44 to 286)
Combined estimate                 70/398 vs. 34/398                     2.28 (1.44 to 3.61)

Figure 15.1. Comparison of Pneumonia Frequency in Stroke Patients between
Dysphagic and Non-Dysphagia

Table 15.9 Relationship Between Aspiration and Pneumonia
        Study          Incidence of Pneumonia Among Patients with   OR (95% CI, fixed effects model)
                                  and without Aspiration
Holas et al. 1994                     8/61 vs. 1/53                       7.85 (0.95 to 65)
Schmidt et al. 1994                   5/26 vs. 1/33                       7.62 (0.83 to 70)
Kidd et al. 1995                      17/25 vs. 2/35                     35.06 (6.69 to 184)
Smithard et al. 1996                  7/20 vs. 12/74                     2.78 (0.92 to 8.42)
Teasell et al. 1996                  10/84 vs. 2/357                      24 (5.15 to 112)
Reynolds et al. 1998                  12/35 vs. 9/68                     3.53 (0.87 to 16.5)
Ding & Logemann                     61/175 vs. 40/203                    1.88 (1.18 to 2.99)
Meng et al. 2000                    3/7 vs. 0/13                           21 (0.90 to 490)
Lim et al. 2001                    5/26 vs. 0/24                         12.53 (0.65 to 240)
Combined estimate               128/468 vs. 67/850                      6.53 (2.91 to 14.64)

15.   Dysphagia                                                                         pg. 13 of 60
Figure 15.2. Comparison of Pneumonia Frequency in Stroke Patients between
Aspirators and Non-Aspirators

 From the pooled results presented
 above (Figures 15.2 & 15.3) the               Dysphagia and aspiration are both
 presence of aspiration was                    associated with an increase in the odds
 associated with a 4.5-fold increased          of developing pneumonia. The risk of
 risk of pneumonia while dysphagia             developing pneumonia appears to be
 (with or without aspiration) was              proportional to the severity of the
 associated with a 3-fold increase in          aspiration.
                                                The risk of developing pneumonia
 Conclusions Regarding the                      following stroke is proportional to the
 Relationship Between Aspiration and            severity of aspiration.
                                             can be assessed, preferably before the
15.6 Non-Instrumental Methods                third day after the stroke. On the
for Screening and Assessment of              other hand, screening describes the
                                             problem in detail, determines the
Dysphagia Following Stroke                   severity of the swallowing problem
Stroke survivors should be screened          and identifies optimal management
for dysphagia as soon as possible after      strategies, including the need for a
acute stroke has been diagnosed and          modified diet or enteral feeding.
emergency treatment has been given           Assessment includes a clinical bedside
and before any oral intake is allowed.       examination and, if warranted by the
Ideally, screening should take place as      clinical signs, an instrumental
soon as the stroke survivor is awake         examination, such as
and alert. Stroke survivors who pass         videofluoroscopy. (Heart and Stroke
the screening are unlikely to have           Foundation of Ontario 2002). Some
significant swallowing difficulties and      common methods for screening and
have a minimal risk of dysphagic             assessment of dysphagia are
complications. Individuals who fail the      described in the following sections.
screen are maintained NPO until they
                                             The Agency for Healthcare Research
15.6.1 Clinical Screening Methods            and Quality published “Evidence
                                             Report/Technology Assessment on

15.   Dysphagia                                                             pg. 14 of 60
Diagnosis and Treatment of                             developed. Most of these screening
Swallowing Disorders in Acute-Care                     tests are comprised of two (or more)
Stroke Patients” in 1999. One of the                   components. Typically, there is some
conclusions reached by this group was                  form of swallowing trial, which is
that no screening tool has yet been                    preceded by a questionnaire or
developed that will accurately detect                  preliminary examination. A description
patients with dysphagia who require                    of the most familiar of these tools is
more extensive testing. Nevertheless,                  presented in Table 15.10.
many screening tools have been

Table 15.10 Description of Screening Tests Used to Identify Dysphagia Post Stroke
  Author/Name of test              Components of test                                Results
                                 Details of validation study
DePippo et al. 1992 44 consecutive patients on a stroke               The sensitivity and specificity of
                    rehabilitation unit with suspected dysphagia      the WST to detect aspiration were
The Burke Dysphagia were studied. Patients were given 3 oz of         76% and 59%, respectively
Screening Test      water from a cup and asked to drink without
                    interruption. Coughing for up to 1 minute
                    after the test or a wet-horse voice was
                    considered abnormal. Patients also received
                    a VMBS study and the results from the 2
                    tests were compared.
Daniels et al. 1997 59 acute stroke survivors were studied. Six       44/59 patients (74.6%) were
                    clinical features-dysphonia, dysarthria,          dysphagic based on VMBS results.
“Any Two”           abnormal volitional cough (includes water-        The presence of 2 clinical features
                    swallowing test), abnormal gag reflex,            correctly distinguished between
                    cough after swallow and voice change after        subjects with normal swallow of
                    swallow were assessed. All subjects               mild dysphagia from those with
                    received a VMBS study in addition to a            moderate or severe dysphagia as
                    clinical exam and water swallowing test.          determined by VMBS examination:
                                                                      Sensitivity: 92%
                                                                      Specificity: 67%
Hinds & Wiles 1998      Standardized questionnaire (11 questions)     The ability of the 11 questions to
                        Timed test of swallowing: subject is given    predict the need for a SLP referral:
“Timed test”            small amount of water from a teaspoon. If     Sensitivity: 0% - 69%
                        successful, 100-150 mL of water is given      Specificity: 62%- 94%
                        with the instruction to drink as quickly as   The ability of the water swallowing
                        possible. A test is considered abnormal if    test to predict the need for a SLP
                        wet hoarse voice or coughing are noted, or    referral:
                        if volume of water consumed are below         Sensitivity: 100%
                        population norms.                             Specificity: 52%
                        115 consecutive subjects with acute stroke.
                        The tool was used to predict the need for
                        SLP intervention.
Logemann et al.         28 items divided into 5 categories:           Aspiration: Throat clearing was
1999                    i) 4 medical history variables                best single predictor.
                        ii) 6 behavioural variables                   Sensitivity:78%
                        iii) 2 gross motor variables                  Specificity: 58%
                        iv) 9 observations from oromotor testing      Oral stage disorder: dysarthria was
                        v) 7 observations during trial swallows       the best single predictor.
                        The tool was designed to identify the         Sensitivity: 64%

15.   Dysphagia                                                                                pg. 15 of 60
                      presence or absence of aspiration, oral    Specificity:75%
                      stage disorder, pharyngeal delay,          Presence of pharyngeal delay:
                      pharyngeal stage disorder.                 being rated as unsafe on at least
                                                                 8/28 swallowing trials was the best
                     202 consecutive patients (34% stroke) were predictor.
                     examined. The results were of the screening Sensitivity: 69%
                     tool were compared with a VMBS exam.        Specificity: 71%
                                                                 Pharyngeal stage swallow disorder:
                                                                 reduced laryngeal elevation was
                                                                 the best single predictor.
                                                                 Sensitivity: 72%
                                                                 Specificity: 67%
Trapl et al. 2007    Preliminary Assessment (vigilance, throat   First group of 50 patients: using a
                     clearing, saliva swallow)                   cut-off score of 14, the sensitivity
The Gugging          Direct swallow ( semisolid, liquid, solid   of GUSS to identify subjects at risk
Swallowing Screen    swallow trials)                             of aspiration: 100%
(GUSS)               Score: 0 (worst) - 20 (no dysphagia)        Specificity: 50%
                     80 acute stroke patients were included.     Second group of 30 patients
                     Results were compared with fibertopic       Sensitivity: 100%
                     endoscopic evaluation.                      Specificity: 69%
Martino et al. 2009  311 stroke patients (103 acute, 208         Prevalence of dysphagia identified
                     rehabilitation) were studied. The tool was  using VMBS: 39%
The Toronto Bedside designed to identify the presence/absence
Swallowing Screening of dysphagia.                               Sensitivity: 91%
Test (TOR-BSST)                                                  Specificity: 67%
                     Items included: voice before, tongue
                     movement, water swallow and voice after.    Reliability (based on observations
                     The results of the screening tool were      from 50 subjects) ICC =0.92 (95%
                     compared with a subset of subjects who      CI: 0.85-0.96)
                     also received a VMBS exam.

                      Scoring: pass-4/4 items; fail ≥1/4 items
Edmiaston et al.      300 acute stroke patients screened by         Prevalence of Dysphagia identified
2009                  nurses within 8 to 32 hours following         using MASA: 29%
USA                   admission.
                                                                    Sensitivity (Dysphagia): 91%
Acute Stroke          Items included: Glasgow Coma Scale score      Specificity: 74%
Dysphagia Screen      <13, presence of facial, tongue or palatal
                      asymmetry/weakness. If no to all 3 items,     Sensitivity (aspiration risk): 95%
                      then proceed to 3 oz water swallowing test.   Specificity: 68%
                      If no evidence of swallowing problems on
                      water swallowing test, then the patient       Inter-rater reliability: 94%
                      passes the screen.                            Test-restest reliability: 92.5%

                      Scoring: pass-4/4 items; fail ≥1/4 items

                    Results were compared with the results of
                    the Mann Assessment of Swallowing Ability,
                    performed by a SPL.
Turner-Lawrence et  A convenience sample of 84 stroke patients Prevalence of Dysphagia identified
al. 2009            (ischemic/hemorrhagic) was included.       by SLPs: 48 (57%)
                    Examinations were conducted by 45 ER
Emergency Physician MDs.                                       Sensitivity: 96%

15.   Dysphagia                                                                          pg. 16 of 60
Dysphagia Screen
                       The two-tiered bedside tool was developed      Specificity: 56%
                       by SLPs.
                       Tier 1 items included: voice quality,          +Likelihood ratio: 2.2
                       swallowing complaints, facial asymmetry,
                       and aphasia.                                   Reliability: Kappa=0.90
                       Tier 2 items included a water swallow test,
                       with evaluation for swallowing difficulty,
                       voice quality compromise, and pulse
                       oximetry desaturation (>or=2%).

                       Patients failing tier 1 did not move forward
                       to tier 2.

                       Patients who passed both tiers were
                       considered to be low-risk. These results
                       were compared with those from a formal
                       assessment by an SLP. Reliability was
                       assessed using a convenience sample of 32
Antonios et al. 2010   150 consecutive patients with acute            Prevalence of Dysphagia identified
                       ischemic stroke were assessed by 2             by SLP using MASA: 54 (36.2%)
Modified Mann          neurologists shortly after admission to
Assessment of          hospital. The results were compared with       Sensitivity: 87% & 93%
Swallowing Ability     the assessments conducted by SLPs using        Specificity: 86% & 84%
(MMASA)                the full MASA.
                                                                      + Predictive Value: 79 & 76%
                       12 of the 24 MASA items were retained
                       including: alertness, co-operation,            + Likelihood Ratio: 5.5 & 6.8
                       respiration, expressive dysphasia, auditory
                       comprehension, dysarthria, saliva, tongue      Reliability: Kappa=0.76
                       movement, tongue strength, gag, volitional
                       cough and palate movement. Maximum
                       score is 100.

In addition to multiple component                       used as both a stand alone screening
tests, stand alone tests can be used to                 method and also as part of a clinical
screen for dysphagia. We examine two                    swallowing screening or assessment.
variations of the water swallowing test                 While the original test required a
in the two tables below.                                patient to swallow 3 oz (90 mL) of
                                                        water, lesser amounts have also been
15.6.2 The Water Swallowing Test                        used. The results of studies, which
                                                        have evaluated this technique, are
The water-swallowing test has also                      detailed in Table 15.11.
been studied extensively. It has been

Table 15.11 Sensitivity and Specificity of the Water-Swallowing Test
 Author/Year                    Methods                                       Outcome
 PEDro score
Garon et al.   100 patients (50% stroke) with confirmed The sensitivity and specificity of the tool to
1995           or suspected dysphagia that required a   identify confirmed aspirators were 54% and
USA            VMBS study as part of clinical           79%, respectively.

15.   Dysphagia                                                                            pg. 17 of 60
No Score       management. All patients were asked to
               drink 3 oz. of water from a cup without
               interruption. Any coughing or throat
               clearing was indicative of an abnormal
               water-swallowing test (WST). The results
               of the 2 methods were compared.
Lim et al.     50 acute stroke patients received a 50 mL    The 50-ml water swallow test had a
2001           water swallowing test (in 10 mL aliquots)    sensitivity of 84.6% and specificity of 75.0%.
Singapore      and a FEES examination. Patients also        The oxygen desaturation test had a sensitivity
No Score       received an oxygen desaturation test.        of 76.9% and specificity of 83.3%. When the
                                                            two tests were combined into one test called
                                                            "bedside aspiration," the sensitivity rose to
                                                            100% with a specificity of 70.8%. Five (10%)
                                                            patients developed pneumonia during their
                                                            inpatient stay. The relative risk (RR) of
                                                            developing pneumonia, if there was evidence
                                                            of aspiration on FEES, was 1.24 (95% CI
                                                            1.03, 1.49).
Chong et al.   50 patients with suspected dysphagia, 65     The WST had a sensitivity of 79.4% and
2003           years or older, who had suffered either a    specificity of 62.5% for the detection of
Singapore      recent or remote stroke. Patients received   aspiration, with a positive predictive value
No Score       a clinical evaluation of swallowing which    (PPV) of 81.8% and a negative predictive
               included a water swallowing test (WST),      value (NPV) of 58.8%. The oxygen
               where patients were asked to drink 50 mL     desaturation test had a sensitivity of 55.9%
               of water in 10mL aliquots, and an oxygen     and a specificity of 100% with PPV of 100%
               desaturation test (desaturation of ≥ 2%      and NPV of 51.6%. When both tests were
               was considered clinically significant) and   combined, a sensitivity of 94.1% and a
               an objective test, fiberoptic endoscopic     specificity of 62.5% were attained, with PPV
               evaluation of swallowing (FEES), where       of 84.2% and NPV of 83.3%. Using the
               episodes of aspiration or penetration of     clinical assessment test, 3 aspirators were
               various food consistencies were noted.       detected who would otherwise have been
               The consistency or results between the       missed if they were assessed with the water
               tests were compared.                         swallow test using thin fluids alone.
Wu et al.      59 stroke outpatients with suspected         55 patients were identified as having some
2004           dysphagia underwent a 100 mL water-          form of swallowing dysfunction on VMBS
Taiwan         swallowing test. Signs of choking or a       examination. An abnormal swallowing speed
No Score       wet sounding voice within 1 minute of        was detected in 47/55 patients. 2 patients
               completing the test were considered          with a normal VMBS result demonstrated
               evidence of an abnormal swallow.             abnormal swallowing speed on the WST. The
               Swallowing speed (< 10 mL/s or ≥ 10          sensitivity and specificity of the test was
               mL/s) was also recorded. The results         85.5% and 50%, respectively. 33 patients
               were compared to a VMBS study.               either aspirated or demonstrated penetration
                                                            on VMBS study. Of these 11 choked on the
                                                            WST, while 3 patients with a normal VMBS
                                                            result, choked on the WST. The sensitivity
                                                            and specificity of the test was 47.8% and
                                                            91.7%, respectively.
Nishiwaki et   61 consecutive stroke patients admitted      Cough/voice change in the water swallowing
al. 2005       to 4 hospitals were assessed for             test was the only variable that was
Japan          dysphagia. Symptoms of oromotor              significantly associated with aspiration on
No Score       functions were evaluated (lip closure,       VMBS examination with sensitivity of 72%
               tongue movement, palatal elevation, gag      and a specificity of 67%.
               reflex, voice quality and motor speech

15.   Dysphagia                                                                            pg. 18 of 60
               function). The water swallowing test
               (using 30 mL of water), saliva swallowing
               test and a VMBS examination were also
               conducted. Factor analysis was used to
               predict dysphagia in patients following

                                                       the onset of swallowing, which is
15.6.3 Swallowing Provocation Test                     identified by visual observation of the
(SPT)                                                  characteristic laryngeal movement,
                                                       and measured with a stopwatch. The
The SPT is a less frequently                           responses to the SPT are classified as
encountered two-stage screening test                   normal or abnormal according to the
that involves the bolus injection of 0.4               induction of the swallowing reflex after
mL and then 2.0 mL of distilled water                  the water injection. A time of seconds
at the suprapharynx through a small                    is used as a cut-off point to
nasal catheter (internal diameter 0.5                  differentiate a normal from an
mm). This manoeuvre elicits an                         abnormal swallow.
involuntary swallow. The latent time is
then timed from the water injection to

Table 15.12 The Water Provocation Test
 Author/Year                    Methods                                          Outcome
 PEDro score
Teramoto &     A retrospective comparative trial of 26       The sensitivity and specificity of first-step SPT
Fukuchi        stroke patients with aspiration pneumonia     for the detection of aspiration pneumonia
2000           and 26 age-matched controls without           were 100% and 83.8%, respectively. Those
Japan          pneumonia were selected to assess the         of the second-step SPT were 76.4% and
No Score       properties of a swallowing provocation        100%, respectively. The sensitivity and
               test (SPT) and a water swallowing test        specificity of first-step WST using 10mL of
               (WST) in detecting aspiration pneumonia       water for the detection of aspiration
               in elderly patients. The normal response      pneumonia were 71.4% and 70.8%,
               to SPT was determined by inducing             respectively. Those of the second-step WST
               swallowing reflex within 3 seconds after      using 30mL of water were 72% and 70.3%,
               water injection into the suprapharynx. In     respectively.
               WST, subjects drank quantities of 10 and
               30mL of water from a cup within 10
               seconds. A test was considered normal if
               the subject drank water without
               interruption and without evidence of
Warnecke et    100 patients with first-ever stroke were      The incidence of endoscopically proven
al. 2008       examined by SPT and fiberoptic                aspiration risk was 81%. The 1st-step SPT
Germany        endoscopic evaluation of swallowing           had a sensitivity of 74% and a specificity of
No Score       (FEES) within 72 hours of stroke onset. A     100%. The 2nd-step SPT had 49% sensitivity
               two-step approach was used. In the first      and 100% specificity.
               step 0.4 ml of distilled water was used. In
               step 2, 2.0 ml was used.

We have presented a variety of                          the detection of dysphagia and
techniques and tools available to aid in                aspiration. Once a patient fails a

15.   Dysphagia                                                                               pg. 19 of 60
screening test and it has been                   orophayngeal dysphagia suggest that
determined that a problem exists,                there is only sufficient evidence to
typically a more comprehensive                   support the value of two tests:
assessment follows, from which                   abnormal pharyngeal sensation and
treatment options are determined.                the 50 mL water-swallowing test. Both
                                                 of these tests assessed only for the
To be clinically useful, screening tests         presence or absence of aspiration.
need to be valid, reliable, easy to use,         Their associated likelihood ratios were
non-invasive, quick to administer (15-           5.7 (95% CI 2.5-12.9) and 2.5 (95%
20 min) and pose little risk to the              CI 1.7-3.7), respectively. Limited
patient. Although many screening                 evidence for screening benefit
tools have been developed it is                  suggested a reduction in pneumonia,
unclear how many of them are used in             length of hospital stay, personnel
institutions beyond those where they             costs and patients.
were developed. Many institutions use
informal processes, or simply restrict
all food and drink until complete                15.6.4 The Bedside Clinical
assessment by an SLP. A wide range               Examination for Assessment of
of sensitivities were reported among             Dysphagia
the tools we reviewed (0% to 100%).
Usually, as sensitivity increased,               Several forms of clinical or bedside
specificity decreased, such that the             swallowing evaluations have been
number of patients who were                      described for the purposes of
incorrectly identified as dysphagic              screening and/or assessment. Some of
increased. Generally screening tools             these methods target specific
with sensitivity > 80%, with a                   functions or tasks, while others
specificity that approaches this figure          evaluate swallowing ability using a
are considered to be both valid and              more comprehensive approach. These
clinically useful. The majority of the           methods may or may not include a
tools presented above do meet these              water-swallowing test. Many of these
criteria.                                        methods have been described
                                                 previously in the section on screening
The results of a systematic review by            and share common features. (Table
Martino et al. (2000), evaluating the            15.13).
screening accuracy of 49 individual
clinical screening tests for

Table 15.13 Components of Various Bedside Techniques to Screen for or Assess
Author/Name of       Components of test
test                 Details of validation study
Smithard et al.      121 stroke patients consecutively          50% of the patients were
1997                 admitted to an urban hospital.             considered to have an unsafe
                     Patients were given an assessment of       swallow based on bedside
Bedside Swallowing   conscious level, head and trunk            evaluation alone.
Assessment           control, breathing pattern, lip closure,   Of these, 20 (16.5%)
                     palate movement, laryngeal function,       patients aspirated on VMBS.
                     gag and voluntary cough (includes          Sensitivity of bedside exam
                     water-swallow test). Patients received     to detect aspiration on
                     both bedside and VMBS (n=94)               VMBS: 47% (SLP); 68%

15.   Dysphagia                                                                  pg. 20 of 60
                    evaluations within 3 days of stroke.       (MD).
                    Both an MD and a SLP each conducted        Specificity: 86% (SLP); 67%
                    the bedside exam.                          (MD)
Mann et al. 2002    128 acute first-ever stroke patients       65 (51%) subjects
                    received both a bedside and VMBS           demonstrated evidence of
Mann Assessment     exam. General examination:                 dysphagia on bedside exam
of Swallowing       Consciousness, cooperation, language       and 28 (21.9%)
Ability             function, verbal/oral praxis,              demonstrated evidence of
                    articulation                               aspiration on VMBS exam.

                    24 items including-                        Using a cut-off score of <180
                    Oral preparation: Control of saliva, lip   the sensitivity and specificity
                    seal, tongue movement/strength, oral       of bedside assessment to
                    preparation, assessment of respiration     detect dysphagia were 71%
                    Oral phase: Gag reflex, palatal            and 72%.
                    movement, oral transit time, bolus
                    clearance                                  The sensitivity and specificity
                    Pharyngeal phase: Pharyngeal               of bedside assessment to
                    control/pooling, laryngeal elevation,      detect aspiration were 93%
                    reflex/voluntary cough, voice quality      and 53%.
                    Includes water swallowing test

                    Scoring for dysphagia:
                    No abnormality ≤178-200
                    Mild ≤ 168-177
                    Moderate ≤ 139-167
                    Severe ≤ 138

15.6.5 Other Methods                           of whom had suffered a stroke) using
                                               VFS as the diagnostic gold standard.
In addition to conventional assessment         Of five voice parameters tested
methods tracheal pH monitoring has             (average fundamental frequency,
also been used experimentally to               relative average perturbation, shimmer
detect drops in pH, which may indicate         percentage, noise-to-harmonic ratio,
aspiration. Clayton et al. (2006)              and voice turbulence index), relative
reported that in 9 of 32 patients              average perturbation most accurately
examined, there was a drop in tracheal         predicted aspiration.
pH following ingestion of acidic foods.
Tracheal pH was monitored by the use           Cervical auscultation of the mechanical
of a sensor, which was inserted into           and/or respiratory components of
the trachea by the crichothyroid               swallowing, lateral cervical soft tissue
membrane. All patients were studied            radiographs and pharyngeal or
following the ingestion of foods which         esophageal manometry have also been
had been considered to be safe on the          used to detect dysphagia (as reviewed
basis of a VMBS examination.                   by Ramsey et al. 2003).

Other forms of clinical assessment             While bedside assessment and other
have been used to detect the presence          non-invasive methods are easy to
of aspiration. Ryu et al. (2004)               perform, these methods have been
recently evaluated voice analysis as a         shown to predict poorly the presence
means to clinically predict laryngeal          of silent aspiration. Smith et al. (2000)
penetration among 93 patients (46%             reported that aspiration cannot be

15.   Dysphagia                                                                  pg. 21 of 60
distinguished from laryngeal                  pudding, bread, and cookies are
penetration using a bedside evaluation,       routinely used. Various aspects of
resulting in the over diagnosis of            oral, laryngeal, and pharyngeal
aspiration and, in some cases,                involvement are noted during the
needless dietary restrictions.                radiographic examination (Table
Therefore, instrumental methods are           15.14). The VMBS study is then
frequently used to directly observe the       followed by a chest x-ray to document
swallowing mechanism.                         any barium, which may have been
                                              aspirated into the tracheobronchial
 Conclusions Regarding Dysphagia              tree.
 Screening and Non-instrumental
 Assessment Techniques                        The VMBS assessment not only
                                              establishes the presence and extent of
 Although a wide variety of screening         aspiration but may also reveal the
 and assessment tests are available for       mechanism of the swallowing disorder.
 use, none have acceptable sensitivity        Aspiration most often results from a
 and specificity to ensure accurate
                                              functional disturbance in the
 detection of dysphagia.
                                              pharyngeal phase of swallowing related
                                              to reduced laryngeal closure or
15.7 Instrumental Methods Used in             pharyngeal paresis. A VMBS study is
the Detection of                              recommended in those cases where
Dysphagia/Aspiration                          the patient is experiencing obvious
                                              problems maintaining adequate
                                              hydration/nutrition, where concern is
15.7.1 VMBS Examination                       expressed regarding frequent choking
                                              while eating, or in the case of
When aspiration is suspected, the             recurrent respiratory infections. Other
videofluoroscopic modified barium             factors such as cognition, recurrent
swallow (VMBS) study is often                 stroke, depression,
considered the "gold standard" in             immunocompromization, and
confirming the diagnosis (Splaingard et       underlying lung disease must also be
al. 1988). A VMBS study examines the          considered. A definitive criterion to
oral and pharyngeal phases of                 determine if a VMBS study is required
swallowing. The patient must have             has yet to be determined in a
sufficient cognitive and physical skills      systematic and scientific manner.
to undergo testing (Bach et al. 1989).        Repeat VMBS studies are usually
The subject is placed in the sitting          conducted at the discretion of the
position in a chair designed to simulate      SLP/MD based on the progress and
the typical mealtime posture. Radio-          prognosis of the individual patient. No
opaque materials of various                   standard schedule for re-assessment
consistencies are tested: barium              exists.
impregnated thin and thick liquids,

15.   Dysphagia                                                           pg. 22 of 60
Table 15.14 Radiological Evaluation During VMBS (from Bach et al. 1989)

Oral Phase
Lips: Closure
Tongue: Anterior and posterior motion with consonants; motion and coordination during
       transport, and manipulation of bolus
Soft palate: Evaluation and retraction with consonants
Jaw: Motion
Oral: Pocketing

Pharyngeal Phase
Swallow: Delay, absence
Peristalsis: Residue in valleculae, pyriform sinuses nasopharyngeal regurgitation

Laryngeal Function
Elevation of larynx
Penetration into laryngeal vestibule
Cough: Presence, delay, effectiveness
Vocal cord function

Post Exam Chest X-Ray
Chronic changes
Presence of barium in valleculae, pyriform sinuses, tracheoboncheal tree, lungs

While VMBS studies can be useful in                      Although VMBS studies are considered
analyzing the anatomic structures                        the gold standard for detection of
during swallowing and detecting silent                   aspiration, other clinical assessment
aspiration, there are some                               techniques, designed to be less
disadvantages: i) The procedure is                       invasive, cheaper and easier to
relatively complex, time consuming                       administer are in current use. Flexible
and resource intensive; ii) there is                     endoscopic examination of swallowing
some exposure to small amounts of                        (FEES), also referred to as fibertopic
radiation; iii) the test is not                          endoscopic evaluation of swallowing,
appropriate for some patients who                        is also recognized as an objective tool
may                                                      for the assessment of swallowing
have difficulty in sitting upright in a                  function and aspiration. The method
chair. The results of the test can also                  has been demonstrated to be safe and
be difficult to interpret and there can                  well-tolerated (Warneke et al. 2009).
be significant variation among                           FEES is a procedure that allows for the
individual raters (Ramsay et al. 2003).                  direct viewing of swallowing function.
                                                         The procedure involves passing a very
15.7.2 Flexible Endoscopic Evaluation                    thin flexible fiberoptic tube through
of Swallowing (FEES)                                     the nose to obtain a view directly
                                                         down the throat during swallowing.
                                                         FEES allows for the full evaluation of

15.   Dysphagia                                                                      pg. 23 of 60
the swallow function as food passes                  the stroke patients, the incidence of
from the mouth into the throat. It is                pneumonia managed by FEESST was
able to identify functional                          significantly lower. The authors
abnormalities that may occur and is                  speculated that one of the reasons for
used in 'practice swallows' to help                  the lower incidence might be due to
determine the safest position and food               the sensory testing component of the
texture to maximize nutritional status               FEES examination, absent from VMBS
and eliminate the risk of aspiration                 evaluation, information which was
and unsafe swallowing. In addition to                used to more effectively guide
assessing the motor components of                    management.
swallowing, FEES can also include a
sensory testing assessment when an                   Rather than attempt to compare the
air pulse is delivered to the mucosa                 accuracy of swallowing abnormalities
innervated by the superior laryngeal                 assessed between VMBS and FEES
nerve. This form of assessment is                    evaluations Leder & Espinosa (2002)
known as flexible endoscopic                         compared the ability of six clinical
examination of swallowing with                       identifiers of aspiration (dysphonia,
sensory testing (FEESST). FEESST was                 dysarthria, abnormal gag reflex,
shown to be a safe technique when                    abnormal volitional cough, cough after
used to assess the swallowing function               swallow, and voice change after
of 500 consecutive subjects. There                   swallow), with FEES to determine the
were only three occurrences of                       a ccuracy of predicting aspiration risk
nosebleeds and no instances of a                     following stroke. Their results suggest
compromised airway. The procedure                    that the ability of the test to correctly
was generally found to be, at worst,                 identify patients not at risk of
mildly uncomfortable (Aviv et al.                    aspiration, was poor using clinical
2000).                                               criteria. Two studies used FEES as the
                                                     gold standard to assess the accuracy
Aviv et al. (2000) compared the                      of either the water-swallowing test
incidence of pneumonia over a one-                   and/or pulse oximetry to detect
year period between patients                         aspiration (Lim et al. 2001, Chong et
managed by VMBS or FEES. Among                       al. 2003).

Table 15.15 Studies Evaluating FEES
 Author/Year                   Methods                                       Outcome
 PEDro score
Aviv 2000      78 outpatients referred for dysphagia      There was no difference in the incidence of
USA            evaluation were assigned to a VMBS         pneumonia between the groups. At the end of
No Score       group to guide swallowing management,      one-year 14 (18.4%) patients whose
               while 61 patients received FEES with       management had been guided by MBS
               sensory testing. The incidence of          developed pneumonia, compared with 6
               pneumonia over a one-year period           (12%) patients in the FEES group (p< 0.20).
               between groups was compared. Patients      However, among 45 stroke patients the
               received feeding tubes, therapy from a     incidence of pneumonia was lower among
               speech-language pathologist based on the   FEES group patients (1/21 vs. 7/24, p<0.05).
               results obtained from the VMBS/FEES test
Leder &        53 consecutive stroke patients referred    The clinical exam correctly identified 19/22
Espinosa       for swallowing assessment were             patients considered at risk for aspiration. The

15.   Dysphagia                                                                           pg. 24 of 60
(2002)         evaluated for the presence of aspiration     clinical exam incorrectly identified 8/27
No Score       using a bedside evaluation which was         patients to be at risk of aspiration. The
               immediately followed by a FEES               sensitivity and specificity of the clinical
               examination. FEES was used as the            assessment were 86% and 30%, respectively.
               diagnostic standard.                         The corresponding positive and negative
                                                            predictive values were 50% and 73%.

15.7.3 Pulse Oximetry

Pulse oximetry has also been                              simultaneously by VFS, among 60
suggested as an alternative to                            patients with dysphagia due to stroke
detecting aspiration, based on the                        and nasopharygeal cancer, while
principle that aspiration of food into                    Collins and Bakheit (1997) reported
the airway leads to bronchospam or                        that pulse oximetry could be used to
airway obstruction, which leads to a                      detect a high proportion of stroke
reduction in oxygen saturation. This                      patients who aspirated on VMBS.
technique is non-invasive, requires
little patient cooperation and is easy                    Age may also be a factor in predicting
to obtain. However, the accuracy of                       oxygen saturation. Rowat et al.
pulse oximetry in detecting aspiration                    (2000) reported that the baseline
is unproven and it remains uncertain                      oxygen saturation among a group of
whether oxygen desaturation can                           stroke patients deemed safe to feed
predict aspiration. Wang et al. (2005)                    orally was significantly lower
reported no significant association                       compared to both hospitalized elderly
between the reduction in oxygen                           patients and young healthy subjects
saturation and aspiration, identified                     (95.7 vs. 96.7 vs. 97.9%, p<0.001).

Table 15.16 Studies Evaluating Pulse Oximetry
 Author/Year                    Methods                                       Outcome
 PEDro score
Collins &      54 consecutive stroke patients with          22 patients demonstrated aspiration on VMBS
Bakheit        swallowing difficulties were studied.        evaluation. Correlation of the pulse oximetry
1997           Patients received a VMBS study and           results with VMBS findings showed that 12
UK             simultaneously had their arterial oxygen     (55%) of the patients who aspirated had a
No Score       saturation measured. The barium meal         significant degree of oxygen desaturation at
               consisted of 150 mL liquid, 3 oz. mousse     the point of swallow/aspiration, but none of
               and biscuit. A drop of 2% in the arterial    the nonaspirators desaturated by 2%. When
               oxygen saturation was considered             the results of oximetry at swallow/aspiration
               clinically significant. Oxygen saturation    and at 2 minutes after swallowing were
               was measured during swallowing, 2            combined, 16 (73%) of the aspirators could
               minutes after the test meal and 10           be identified by this method, and 4 (13%) of
               minutes after the VMBS study was             the nonaspirators also had a significant
               completed.                                   oxygen desaturation. In total, 44 patients
                                                            (81.5%) were accurately predicted as
                                                            aspirators or nonaspirators ( =0.61, P<.001).
                                                            Prediction was better for males compared to
                                                            females. The sensitivity and specificity of
                                                            pulse oximetry were 73% and 87%,

15.   Dysphagia                                                                           pg. 25 of 60
Sellars et al.   Six patients (4 with stroke) with          4 patients demonstrated aspiration of VMBS.
1998             established dysphagia underwent both       Of these, 2 exhibited significant 02
UK               VMBS evaluation with simultaneous          desaturation.
No Score         oxygen saturation monitoring. Decline in
                 02 saturation of 4% from baseline was
                 considered clinically significant.
Sherman et       46 (16 with stroke) with swallowing      12/46 patients (6 with stroke) aspirated on
al. 1999         difficulties underwent VMBS evaluation   VMBS. Patients who aspirated had a
USA              with simultaneous oxygen saturation      significantly greater decline in oxygen
No Score         monitoring (with a 5-6 second sampling   saturation compared to those who did not
                 interval).                               aspirate. The lowest O2 saturation value
                                                          among patients who aspirated was 81%
                                                          compared with 92% among patients who did
                                                          not aspirate/penetrate.
Smith et al.  53 consecutive patients with acute stroke 15/53 patients aspirated on VMBS
2000          received a bedside evaluation, pulse        examination. The SN, SP, PPV and NPV for
UK            oximetry and a VMBS evaluation of           pulse oximetry to identify aspiration were
No Score      swallowing. The sensitivity (SN),           87%, 39%, 36% and 88%, respectively.
              specificity (SP), positive predictive value
              (PPV) and negative predictive value (NPV)
              were calculated for both the bedside
              evaluation and pulse oximetry.
Wang et al.   60 patients (27 with stroke) received both 23/60 patients demonstrated aspiration on
2005          oxygen saturation and VMBS evaluation. VMBS examination. Of these patients 9
Taiwan        Oxygen saturation was monitored for 5       displayed significant oxygen desaturation (a
No Score      minutes before and for 5 minutes after      drop of > 3% was considered significant. Of
              the VMBS evaluation.                        the 37 patients who did not demonstrate
                                                          aspiration on VMBS, 15 had an episode of
                                                          oxygen desaturation. The sensitivity and
                                                          specificity were 39.1% and 59.4%,
                                                          respectively. The positive and negative
                                                          predictive values were 37.5% and 61.1%,
                                                          respectively. The positive likelihood ratio was
Ramsey et al. 189 stroke patients received a bedside      15 (28%) demonstrated aspiration on VMBS.
2006          swallowing assessment (BSA), pulse          Of these, 2% destauration was seen in 5
UK            oximetry and VMBS (n=54) studies. Two (33.3%) of these patients and in 2 (13.3%)
No Score      cut-points were selected to determine the when >5% threshold was used. 7/15
              presence/absence of oxygen destauration patients (47%) with demonstrated aspiration,
              (>2% and >5%).                              failed the BSA. The sensitivity and specificity
                                                          associated with >2% destauration were 33%
                                                          and 62% and were 13% and 95% for an
                                                          oxygen desaturation threshold of >5%.

Although pulse oximetry is a quick and                  above studies will attest to.
non-invasive method to detect
aspiration following stroke, its                        15.8 Management of Aspiration
association with oxygen desaturation
have been inconclusive. Generally, its
                                                        Post Stroke
performance when measured against
VMBS studies has been poor as the                       As mentioned previously, the VMBS
low sensitivities/specificities from the                study is still considered the "gold
                                                        standard" in the diagnosis of
15.   Dysphagia                                                                           pg. 26 of 60
aspiration. Those patients who have             Conclusions Regarding Instrumental
difficulty with high volumes of thin            Methods to Detect Dysphagia Post
liquids are considered to be at mild to         Stroke
moderate risk of aspiration. In these
cases oral feedings are regarded as             The VMBS study is considered the gold
appropriate. Before deciding if a               standard to detect silent aspiration.
patient is a candidate for oral feeding,        Other methods such as FEES and pulse
                                                oximetry are also in use.
factors such as the patient's
respiratory status, the effectiveness of
                                              15.8.1 Management Strategies for
airway clearance along with the type
and amount of aspirate must first be
considered (Bach et al. 1989).
                                              The Heart and Stroke Foundation
Aspirating more than 10% of the test
                                              Dysphagia Guidelines noted that, “a
bolus is generally considered an
                                              well coordinated care plan can
indication for non-oral (ie. nasogastric,
                                              minimize the development of
gastrostomy, jejunostomy tube)
                                              dysphagic complications, reduce
feedings; however, the actual risks
                                              length of hospital stay in acute-care
present with oral feedings for this
                                              facilities and expedite access to
group of patients have not been fully
                                              specialized rehabilitation centers.
established. Determining whether the
                                              Dysphagia management has the
patient actually aspirates more or less
                                              following goals:
than 10% of the test bolus is, as
                                              • Meeting the nutrition and hydration
mentioned previously, an inexact
                                              requirements of the stroke survivor.

• Preventing aspiration-related               A speech-language pathologist should
complications.                                regularly monitor the status of
• Maintaining and promoting                   individuals with dysphagia to ensure
swallowing function as much as                that the management strategies
possible.                                     employed remain appropriate,” (Heart
                                              and Stroke Foundation of Ontario
Dysphagia management strategies               2002).
include the following:
• Modifying food and fluid textures to        15.8.2 Best Practice Guidelines for
increase safety of oral intake.               Managing Dysphagia
• Using low-risk feeding practices and
compensatory strategies to prevent            Best practice guidelines for managing
complications such as aspiration and          dysphagia were developed by a
choking.                                      consensus committee sponsored by
• Monitoring oral intake to prevent           the Heart and Stroke Foundation of
dehydration.                                  Ontario (2002). These are
• Supplementing the diet to maintain          summarized in Table 15.17.
adequate nutrition.
• Using enteral feeding for individuals
who are unable to swallow.
• Implementing swallow therapy to
rehabilitate specific physiological
swallowing impairments.

15.   Dysphagia                                                           pg. 27 of 60
Table 15.17         Best Practice Guidelines for Managing Dysphagia Post-Stroke (HSFO
      •   Maintain NPO until swallowing status is determined
      •   Regular oral care, with minimum of water to limit build-up of bacteria
      •   Screen for swallowing status by trained team member once awake and alert
      •   Screen for risk factors of poor nutrition early by trained team member
      •   Swallowing assessment necessary for all those who fail swallowing screen
      •   Swallowing assessment by speech-language pathologist to:
               -   assess ability to swallow
               -   determine swallowing complications
               -   identify associated factors which may be compromising swallowing and nutrition
               -   recommend appropriate individualized management program, including appropriate
               -   monitor hydration status
      •   Where appropriate feeding assistance or mealtime supervision by individuals trained in low-
          risk feeding strategies
      •   Assess nutrition and hydration status and needs of those who fail screening; reassess regularly
      •   Education of patient and family into follow-up upon discharge
      •   Consider the wishes and values of the patient and family concerning oral and non-oral
          nutrition; provide information to allow informed choices.

 Conclusions Based on Best Practice
 Guidelines for Managing Dysphagia                         There is consensus (Level 3) opinion
                                                           that a dietician should assess the
 There is consensus (Level 3) opinion                      nutrition and hydration status of all
 that acute stroke survivors should be                     stroke patients who fail swallowing
 NPO until swallowing ability has been                     screening.
                                                            All stroke survivors should remain
 There is consensus (Level 3) opinion                       NPO until a trained assessor has
 that a trained assessor should screen                      assessed swallowing ability.
 all acute stroke survivors for
 swallowing difficulties as soon as they
 are able.                                                  Feeding assistance should be
                                                            provided by an individual trained in
 There is consensus (Level 3) opinion                       low-risk feeding strategies where
that a speech and language pathologist                      appropriate,
should assess all stroke survivors who
fail swallowing screening and identify
                                                            Following a failed screening
the appropriate course of treatment.
                                                            assessment, all patients should be
                                                            assessed by a Speech-Language
 There is consensus (Level 3) opinion
                                                            Pathologist and an appropriate
 that an individual trained in low-risk
                                                            management plan be initiated.
 feeding strategies should provide
 feeding assistance or supervision to
 stroke survivors where appropriate.

15.   Dysphagia                                                                             pg. 28 of 60
15.8.3 Dysphagia Screening Protocols

Two studies have evaluated whether                    pneumonia. One study used a
the implementation of dysphagia                       historical control while the other used
screening protocols resulted in a                     a concurrent control group (Table
reduction in the incidence of                         15.18).

Table 15.18 Dysphagia Screening Protocols
 Author/Year                   Methods                                       Outcome
 PEDro score
Odderson et    The incidence of pneumonia was assessed     The percentages of patients who developed
al. 1995       in a single institution before the          pneumonia before the pathway was
USA            implementation of a dysphagia screening     developed, during the first year of the
No Score       protocol, during the first year after its   pathway and during the second year of the
               implementation and during the second        pathway were 6.7%, 4.1% and 0%,
               year.                                       respectively.
Hinchey et al. 15 acute care hospital sites were           6 sites had a formal dysphagia screen. Their
2005           surveyed to determine whether they had      adherence rate was 78% compared with 57%
USA            an established dysphagia screening          at sites with no formal screen. The
No Score       protocol in place and to establish the      pneumonia rate at sites with a formal
               adherence level. The incidence of           dysphagia screen was 2.4% versus 5.4%
               pneumonia between institutions which        (p=0.0016) at sites with no formal screen.
               had/ did not have formal screening in       There was no difference in median stroke
               place was compared.                         severity (5 versus 4; P=0.84) between the
                                                           sites with and without a formal screen.

 There is evidence from two studies                   dysphagia who are fed are
 that the initiation of a dysphagia                   approximately 20 times more likely to
 screening program can help to                        develop pneumonia than those who
 reduce the incidences of pneumonia,                  feed themselves (Langmore et al.
 presumably through earlier detection                 1998). Therefore, if dysphagic
 and subsequent management of                         individuals cannot feed themselves
 swallowing difficulties.                             independently, hand- over-hand
                                                      support should be provided from an
 Conclusions Regarding the Benefits of                eye level position. If full feeding
 Dysphagia Screening Protocols                        assistance is necessary, it should be
                                                      provided using low risk feeding
 There is limited (Level 2) evidence that             strategies.
 dysphagia screening protocols can
 reduce the incidence of pneumonia.                   Routine use of low-risk feeding
                                                      strategies can prevent serious health
15.8.4 Low-Risk Feeding Strategies
                                                      problems and improve the quality of
for Dysphagia
                                                      the experience for the person being
                                                      fed. All health care professionals
The Heart and Stroke Foundation                       involved in feeding dysphagic
Dysphagia Guidelines noted that,                      individuals should also be able to deal
“Stroke survivors should be                           with emergencies, such as choking,
encouraged and assisted to feed                       which may occur during feeding.”
themselves. Individuals with

15.   Dysphagia                                                                          pg. 29 of 60
                                                        Guidelines for low-risk feeding
                                                        practices are summarized in Table

Table 15.19 Heart and Stroke Foundation of Ontario Guidelines for low-risk feeding
practices (2002)
•Check the food tray to ensure the correct diet type has been provided.
•Ensure the environment is calm during meals and minimize distractions.
•Position the stroke survivor with the torso at 900 angle to the seating plane, aligned in mid-position with
the neck slightly flexed.
•Support the stroke survivors with pillows if necessary.
•Perform mouth care before each meal to remove bacteria that have accumulated on the oral mucosa.
•Feed from a seated position, so that you are at eye level with the stroke survivor.
•Do not use tablespoons. Use metal teaspoons, never plastic for feeding individuals with bite reflexes.
•Use a slow rate of feeding and offer a level teaspoon each time.
•Encourage safe swallowing of liquids by providing them with wide-mouth cup or glass or in a cut-down
nosey cup, which helps prevent the head from flexing backward and reduces the risk of aspiration. Some
individuals may benefit from drinking through a straw.
•Ensure that swallowing has taken place before offering any additional food or liquid.
•Observe the stroke survivor for any signs or symptoms of swallowing problems during and for 30
minutes after the meal.
•Perform mouth care after each meal to ensure that all food debris is cleared from the mouth.
•Position the patient comfortably upright for at least 30 minutes after each meal to promote esophageal
clearance and gastric emptying and to reduce reflux.
•Monitor the oral intake of the stroke survivor with dysphagia: note any food items that are not
consumed and ensure that intake is adequate, especially important in individuals receiving a thickened-
liquid diet.
•Document the patient’s intake, any changes in swallowing status and any self-feeding problems.

 Conclusions Regarding Feeding
 Strategies in Dysphagia

 There is limited (Level 2) evidence that
 individuals with dysphagia should feed
 themselves to reduce the risk of

 For patients who require assistance to
 feed, there is a consensus (Level 3)
 opinion that low-risk feeding strategies
 by trained personnel should be

  Individuals with dysphagia should
  feed themselves whenever possible.
  When not possible, low-risk feeding
  strategies are needed.

15.   Dysphagia                                                                             pg. 30 of 60
                                                        dysphagia following stroke. The review
15.9 Specific Interventions to                          included 6 studies, including an
Manage Dysphagia                                        abstract and unpublished data,
                                                        assessing how and when to feed, oral
                                                        supplementation and how and when to
Previous Reviews                                        treat. Although few studies were
A Cochrane systematic review (Bath et                   available, the following conclusions
al. 1999) evaluated the benefit of                      were reached.
different management strategies for

Table 15.20 Results From Cochrane Review Evaluating Treatments for Dysphagia

Enteral feeding: Based on the results from 2 studies percutaneous endoscopic gastrostomy (PEG) was
associate with lower case fatality rates, treatment failures and improved nutritional parameters,
compared with nasogastric (NG) tube feeding.

Timing of feeding: No studies

Nutritional supplementation: Based on the results from a single study this intervention was associated
with improved energy and protein intake.

Fluid supplementation: Based on the results of a single trial there was no evidence of a benefit in
reducing the time to resolution of dysphagia.

Swallowing therapy: Based on the results from two studies there no evidence of a reduction in end-of-
trial dysphagia.

Drug therapy: Based on the result of a single trial, there was no evidence of a decrease in mortality or
frequency of dysphagia associated with nifedipine treatment.

A recent systematic review (Foley et
al. 2008) also evaluated the efficacy of               15.9.1 Dietary Modifications
a broader range of dysphagia
treatments including: texture-modified                 Dysphagia diets have three purposes:
diets, general dysphagia therapy                       1) to decrease the risk of aspiration, 2)
programmes, non-oral (enteral)                         to provide adequate nutrients and
feeding, medications, and physical and                 fluids, and 3) to provide a progressive
olfactory stimulation. In this review, 15              approach to feeding based on
RCTs were identified. In contrast with                 improvement or deterioration of
the findings of the Cochrane review,                   swallowing function (Bach et al. 1989).
there was evidence that nasogastric                    No single dysphagia diet exists. Diets
tube feeding was not associated with a                 include modified food and liquid
higher risk of death compared to                       textures (HSFO, 2002). Special diets
percutaneous feeding tubes. General                    are based upon four distinct
dysphagia therapy programmes were                      consistencies: thick fluids, pureed,
associated with a reduced risk of                      minced and soft chopped. A dysphagia
pneumonia in the acute stage of                        soft diet excludes all hard, small and
stroke.                                                stringy food particles (Bach et al.
                                                       1989). There are three consistencies
                                                       of meat in the soft diet; soft chopped,

15.   Dysphagia                                                                            pg. 31 of 60
minced and ground. A pureed diet has        frequent cueing may be necessary in
the consistency of pudding and is           these cases (Milazzo et al. 1989).
generally easier to swallow than a
more regular diet (Veis and Logemann        The restrictions associated with a diet
1985). However, the risk of aspiration      of thick fluids can eliminate all thin
of pureed food was recently reported        liquids. Alternatives to thin liquids
by Perlman et al. (2004). 204 stroke        such as jelled water or liquids may be
patients were divided into six groups       required. There is some evidence that
based on the results of                     dietary modifications may reduce the
laryngopharyngeal sensory testing,          incidence of aspiration pneumonia
assess by flexible endoscopic               (Groher 1987) although it has not
evaluation. No patients with both           been definitively established as to what
normal sensation and pharyngeal             effect the mode of feeding has on the
squeeze aspirated pureed consistency        rate of respiratory infection.
foods. The percentage of aspirators
increased to 67% in patients with           The Heart and Stroke Dysphagia
moderately decreased sensation and          Guidelines noted “Diet texture
absent motor function. The results of       modification, however, can reduce an
this study suggest that motor strength      individual’s enjoyment of food,
may be more important than sensory          resulting in decreased oral intake. This
impairment in the prediction of             can rapidly lead to dehydration and
aspiration.                                 eventually to malnutrition. Also, the
                                            use of starch-based food thickeners
Over time, particularly in the earlier      increases carbohydrate intake, which
stages following stroke, changes to the     may produce a nutritional imbalance if
diet can be made as the patient's           the diet is not carefully monitored.
dysphagia improves and the risk of          Controlling dietary carbohydrates is
aspiration lessens. Progression can be      especially important in individuals with
determined by clinical swallowing           diabetes. It is therefore critical to
assessments unless the patient is a         consult a dietitian to ensure that the
"silent aspirator", detectable only on      modified diet is nutritionally adequate
VMBS study, in which case the clinical      and appropriate, and to consult the
examination must be considered              stroke survivor or substitute decision-
unreliable. A repeat VMBS study may         maker to ensure that the modified diet
be needed in these cases in order to        is as appealing as possible” (Heart and
guide management. Special                   Stroke Foundation of Ontario 2002).
techniques such as compensatory head
and neck postures (Logemann 1983),          Avoidance or careful regulation of thin
double swallowing or coughing after         liquids is a common dietary
swallowing (Horner et al. 1988b) may        modification, as this food consistency
be employed. Many stroke patients,          is the most likely to be aspirated. Thin
especially those with right hemispheric     fluids are poorly manipulated in transit
lesions, are very impulsive and may         through the oral-pharynx. Severely
attempt to eat and swallow at too fast      dysphagic patients are often managed
a rate. Finestone et al. (1998)             initially by enteral tube feedings and
documented a case in which a man,           progress to the re-introduction of oral
post stroke died following airway           feeding, typically beginning with a
obstruction caused by a food bolus.         pureed diet. Eventually patients are
Therefore close supervision with            allowed thin liquids when it has been

15.   Dysphagia                                                          pg. 32 of 60
established that the patient can                      CI 3.6-14.3), Diuretic usage
successfully swallow without                          augmented the risk; an aspirating
aspirating. Currently no randomized                   patient concurrently taking a diuretic
controlled studies have demonstrated                  for hypertension or management of
whether these modified diets influence                congestive heart failure was 20 times
outcome although a large multi-                       more likely to experience dehydration
centered trial has been completed and                 (OR 19.8, 95% CI 3.0-211).
the results are pending publication
(Dennis 1997).                                        In a recent study, Diniz et al. (2009)
                                                      examined 61 acute stroke patients for
Although thickened fluids may help to                 signs of aspiration after receiving both
reduce the risk of aspiration and                     thin liquids and pudding-like feeds
associated morbidity, Finestone et al                 using nasoendoscopy. Aspiration
(2001) reported that patients                         occurred in only 3 patients with the
restricted to thickened fluids not drink              spoon-thick consistency vs. 21 with
sufficient quantities to meet their fluid             the liquid consistency (relative
needs and are at risk for dehydration.                risk=0.13; 95% confidence
Patients receiving dysphagia diets                    interval=0.04-0.39; P<.001). There
along with texture-modified solids                    were no episodes of laryngeal
received only 43% of their estimated                  penetration with pudding-like fluids
fluid requirement over the first 21 days              and 8 incidences with thin liquid.
post stroke, while in hospital. Although              Patients in this study all had feeding
dietary modifications were not                        tubes in situ. However, Leder & Suiter
specifically addressed, Churchill et al.              (2009) reported than the placement of
(2004) found that dysphagic patients                  NG feeding tubes did not increase the
had a higher risk of becoming                         risk of aspiration for liquid or pureed
dehydrated, defined as a peak blood                   food consistencies. This study included
urea nitrogen (BUN) ≥ 45. The odds                    dysphagic patients with a broad range
ratio (OR) associated with dehydration                of etiologies, including stroke. The
was 4.2 (95% CI 2.1–8.3) among                        sample size was large (n=1,260).
patients admitted for inpatient stroke
rehabilitation, and was even higher for               Dietary management is often directed
patients with aspiration, detected                    by the results of the VMBS studies.
through videofluoroscopic                             Studies examining the efficacy of fluid
examinations and presumed to be on a                  modifications are presented in Table
texture-modified diet (OR: 7.2; 95%                   15.21.

Table 15.21 Studies of Dietary Modifications in Dysphagia
  Author/Year                   Methods                                      Outcome
 PEDro score
Groher et al.   56 stroke patients with chronic            Study group had fewer occurrences of
1987            dysphagia, on a pureed diet prior to       aspiration pneumonia (5 vs. 28, p<0.05). (If
USA             study and at least 1 month post            persons who developed pneumonia on more
3 (RCT)         resolution of aspiration pneumonia were    than one occasion are ignored the incidence
                randomized to receive: i) a soft           of pneumonia was still lower in the study
                mechanical diet and thickened liquids or   group: 4 vs. 18).
                ii) pureed foods and thin liquids.
                The recurrence of aspiration pneumonia
                over a 6-month period was assessed.
Garon et al.    20 dysphagic stroke patients were          No patient in either group developed

15.   Dysphagia                                                                          pg. 33 of 60
1997            randomized to receive either a regular       dehydration or pneumonia within the 30- day
USA             dysphagia diet including thickened fluids    trial period, or required intravenous fluids.
5 (RCT)         (control group) or to a dysphagia diet       There were no significant differences in total
                which allowed the inclusion of unlimited     fluid intake between the groups. However,
                amounts of water (study group) between       patients in the study group drank significantly
                meals.                                       less thickened fluid compared to patients in
                                                             the control group.
Goulding &      46 dysphagic inpatients were                 Higher viscosity of fluid prepared using
Bakheit 2000    randomized to receive thickened fluids       subjective assessment. There were no
UK              prepared using conventional subjective       significant differences in the incidence of
6 (RCT)         assessment of viscosity or fluids            aspiration between the groups. Strong
                thickened with the aid of viscometer for     correlation between increased viscosity and
                7 days.                                      portion of thickened fluid that was not
Perlman et al. 204 dysphagic patients underwent              Sensation       Motor Function Aspiration (%)
2004           assessment of swallowing function and         Normal                Normal          0
USA            sensory evaluation with flexible              Normal                Absent         14
No Score       endoscope. Patients were then divided         Mod decrease           Normal        0
               into 3 groups, with normal, moderate          Mod decrease           Absent        67
               and severe sensory deficits. Each group       Severe/absent          Normal         6
               was divided into those with                   Severe/absent          Absent        40
               impaired/normal pharyngeal squeeze.
               Patients were then tested for aspiration
               following a pureed food bolus.
Diniz et al.   61 patients, 19 with acute stroke             Aspiration occurred in 24 patients. A higher
2009           received a trial of either liquid or spoon-   proportion of patients aspirated with liquid
Brazil         thick liquids in random order and were        samples (3 vs. 21, p<0.001). There was no
6 (RCT)        assessed for evidence of dysphagia            evidence of penetration among patients given
               (penetration, aspiration or residue) using    spoon-thick liquids compared with 8 instances
               nasoendoscopy. Patients also underwent        following liquid challenges (p<0.006).
               a clinical examination and bedside
               clinical assessment.

 Conclusions Regarding Dietary                             There is moderate (Level 1b) evidence
 Modifications                                             that thickened fluids result in fewer
                                                           episodes of aspiration and penetration
 There is consensus (Level 3) opinion                      compared with thin fluids among
 that dysphagic stroke patients, who are                   dysphagic individuals following stroke.
 considered safe with oral intake require
 diets with modified food and liquid                         Dysphagic stroke patients should be
 textures. Although dietary                                  provided with an appropriate modified
 modifications have been used to help                        diet, after consultation with a dietitian.
 reduce the risk of aspiration and their
 consequences following stroke, the                     15.9.2 Swallowing Treatment
 evidence in support of their use is
 lacking. Further research is needed in
 this area.                                             Four studies have examined the effect
                                                        of formal dysphagia therapy on a
 There is limited (Level 2) evidence that               variety of outcomes. Dysphagia
 dysphagia diets reduce the incidence                   therapy usually involves a combination
 of aspiration pneumonia.                               of approaches, including exercises
                                                        aimed at strengthening muscles, and
                                                        improving movement and

15.   Dysphagia                                                                              pg. 34 of 60
coordination. Possible exercises may                      found a trend towards statistical
include the Mendelsohn maneuver                           significance when examining the
(the patient hold the larynx up, either                   impact of two levels of dysphagia
using the muscles of the neck or with                     treatment programs (low and high
the hand, during the swallow for an                       intensity) on decreasing the need for a
extended period of time), the Masako                      modified diet. Compared to usual
maneuver (patient protrudes tongue                        care, patients who received instruction
and then swallows), Shaker exercise                       on compensatory swallowing
(                       strategies, swallowing exercises and
360.htm), and gargling, among                             regular re-evaluation of dietary
others. Other strategies include                          modifications were more likely to have
postural changes (head turn and chin                      returned to an unmodified diet at six
tuck postures) and multiple swallows.                     months. Odderson et al. (1995) in an
These therapies are usually provided                      unrated retrospective study found the
in addition to dietary modifications.                     introduction of a stroke program with
                                                          dysphagia therapy improved
DePippo et al. (1994) conducted the                       dysphagia-related outcomes. Lin et al.
only RCT, which demonstrated no                           (2003) also reported improvements in
benefit of formal dysphagia therapy.                      various nutrition parameters and
However, the two-week treatment                           choking frequency among patients
period may have been too short to                         who participated in a swallowing
actually demonstrate a significant                        training program.
difference. Carnaby et al. (2005)

Table 15.22 Studies of Dysphagia Therapy Post Stroke
  Author/Year                      Methods                                        Outcome
 PEDro score
De Pippo et al.   115 patients randomized to receive either     Up to 1 year follow-up revealed no
1994              one formal dysphagia treatment session        significant differences between the 3 groups
USA               and choice of modified-texture diet, one      in the incidence of pneumonia, dehydration
5 (RCT)           dysphagia session with prescribed             No deaths were reported. Only one instance
                  texture-modified diet or daily intervention   of recurrent upper airway obstruction.
                  by SLP and prescribed diet.
Odderson et       124 patients with non-hemorrhagic stroke      48 (39%) patients were diagnosed with
al. 1995          admitted to an urban community hospital.      dysphagia on admission. No incidences of
USA               Within 24 hours of admission, patients        aspiration pneumonia were reported. The
No Score          received a clinical swallowing evaluation     year prior to the introduction of the
                  and received appropriate dysphagia            pathway, 6.7% of patients developed
                  interventions if required, as per the         aspiration pneumonia. The first year the
                  protocol of a recently implemented clinical   pathway was introduced, 4.1% of patients
                  pathway. The incidences of aspiration         developed aspiration pneumonia. Patients
                  pneumonia, LOS and outcome disposition        without dysphagia had a shorter LOS and
                  were recorded (criteria for defining          were more likely to be discharged to the
                  pneumonia was not reported). Functional       community. Patients who passed the initial
                  outcome was assessed using FIM.               swallowing screen had higher FIM scores
                                                                compared to those who failed.
Lin et al.        A quasi-experimental parallel, cluster        The results of between group comparisons
2003              design study that recruited 61 patients       on change scores (pre-test, post test)
Taiwan            (2:1) from 7 long-term care facilities to     showed statistically significant
No Score          receive either swallowing training or no      improvements favouring the treatment
                  therapy (Patients received therapy            group for: swallowing function (incidence of

15.   Dysphagia                                                                               pg. 35 of 60
               following data collection). Swallowing             coughing/choking, volume/second
               training consisted of direct therapies             swallowed, volume per swallow),
               (compensatory strategies, diet                     neurological examination and nutrition
               modification, environmental arrangement,           parameters (mid-arm circumference and
               the Mendelssohn manoeuvre, supraglottic            weight)
               swallowing and effortful swallowing) and
               indirect therapies (thermal stimulation,
               oral motor and lingual exercises and were
               provided 30 min/days 6 days/week x 8
Carnaby et al. 306 patients with clinical dysphagia               Of patients randomly allocated usual care,
2006           admitted to hospital with acute stroke             56% (57/102) survived at 6 months free of
USA            were randomly assigned to receive usual            a modified diet compared with 64%
8 (RCT)        care (n=102), standard low-intensity               (65/102) allocated to standard (low-
               intervention (n=102), or standard high-            intensity) swallowing therapy and 70%
               intensity intervention and dietary                 (71/102) patients who received high-
               prescription (n=102). Treatment                    intensity swallowing therapy. Compared
               continued for up to a month. The primary           with usual care and low-intensity therapy,
               outcome measure was survival free of an            high-intensity therapy was associated with
               abnormal diet at 6 months                          an increased proportion of patients who
                                                                  returned to a normal diet (p=0.04) and
                                                                  recovered swallowing (p=0.02) by 6
                                                                  months. Results also presented in Figure

            Figure 15.3 Percentage of Patients Achieving Normal
                             Diets at 6 M onths

                                 sit   y                si ty                 sit y
                            sten                  nt en              Inte n
                  U sual In                L ow I               High

 Conclusions Regarding Dysphagia                                There is moderate (Level 1b) evidence
 Therapy                                                        that while a one-month dysphagia
                                                                intervention program does not improve
 There is moderate (Level 1b) evidence                          the likelihood of returning to a normal
 that a short course of formal dysphagia                        diet by six months, it may reduce the
 therapy does not alter clinical                                likelihood of chest infections and death
 outcomes.                                                      or institutionalization.

15.   Dysphagia                                                                                pg. 36 of 60
                                                       weeks, a gastrostomy or jejunostomy
  A short course of formal dysphagia                   feeding tube may be indicated,” (Heart
  therapy may not alter clinical                       and Stroke Foundation of Ontario
  outcomes.                                            2002). Although enteral feeding tubes
                                                       have been shown to deliver adequate
15.9.3 Non-Oral Feedings                               nutrition and hydration to stroke
                                                       survivors, and can improve indicators
Non-oral or tube feeding in neurogenic                 of nutritional status, their use has
aspiration has become a well-                          been associated with some medical
established rehabilitation practice. The               complications, most notably, aspiration
Heart and Stroke Dysphagia Guidelines                  pneumonia (Finestone et al. 1995,
state, “Enteral feeding is                             2001, James 1998). However, the
recommended if a swallowing                            association between enteral feeding
assessment indicates high-risk                         and the subsequent development of
dysphagia or inability to meet                         pneumonia remains unclear, tube
nutritional needs orally. Enteral                      feeding has been identified as both
feeding should be considered after a                   protective and a risk factor for
stroke survivor has been NPO for 48                    pneumonia. Table 15.23 presents two
hours and implemented within 3-4                       studies, which have investigated this
days. If dysphagia is severe and                       relationship.
expected to last for more than 6

Table 15.23 Studies which Examine the Risk of Aspiration Pneumonia Associated
with Enteral Feeding
 Author/Year/                   Methods                                       Outcome
 PEDro score
Nakajoh et al. The incidence of pneumonia was              The incidence of pneumonia was significantly
2000           prospectively analyzed for 1 year in        higher in patients with oral feeding than in
Japan          three groups of post-stroke patients on     those with tube feeding (54.3% vs. 13.2%, P
No Score       the basis of the following clinical         < 0.001). In bedridden patients with tube
               conditions: oral feeding without            feeding, the latency of response was longer
               dysphagia (n = 43); oral feeding with       than 20 sec and no patient coughed at the
               dysphagia (n = 48); and nasogastric         highest concentration of citric acid. The
               tube feeding with dysphagia (n = 52).       incidence of pneumonia was 64.3% in such
               The incidence of pneumonia in               patients. The state of protective reflexes had a
               bedridden patients with nasogastric         significant relation to the incidence of
               tube feeding (n = 14) was also              pneumonia. Feeding tube placement may have
               studied. Pre-study, the swallowing and      a beneficial role in preventing aspiration
               cough reflexes of each patient were         pneumonia in mildly or moderately disabled
               measured. The swallowing reflex was         post-stroke patients with attenuated protective
               evaluated according to latency of           reflexes.
               response, which was timed from the
               injection of 1 mL of distilled water into
               the pharynx through a nasal catheter
               to the onset of swallowing.
Dziewas et al. Over an 18-month period, 100                Pneumonia was diagnosed in 44% of the tube
2004           consecutive acute stroke patients who       fed patients. All pneumonias occurred while
Germany        were fed by a naso-gastric feeding          the tube was in situ. Most patients acquired
No Score       tube because of dysphagia were              pneumonia on the second or third day after
               prospectively evaluated.                    stroke onset. Patients with pneumonia more

15.   Dysphagia                                                                              pg. 37 of 60
                                                       often required endotracheal intubation and
                                                       mechanical ventilation than those without
                                                       pneumonia. Independent predictors for the
                                                       occurrence of pneumonia were a decreased
                                                       level of consciousness and severe facial palsy.
Marum & Lim 122 patients admitted to a geriatric       90 patients were recommended for non-oral
2005           ward (75% with stroke) were assessed feeding. Of these, 64 agreed and 26 refused
Singapore      by a SLP. Following assessment          and were fed orally. 32 patients were deemed
               patients were recommended to have       safe with an oral, modified diet. There were 14
               either oral feeding with modified diet  cases of aspiration pneumonia confirmed using
               or nasogastric tube feeding. The        pre-defined criteria during the 2-month follow-
               incidence of aspiration pneumonia       up, resulting in death in 5 cases. 12 of these
               among patients on oral feeding,         cases were reported among patients fed by an
               nasogastric (NG) tube feeding and       NG tube, 2 in patients who refused NG tubes
               patients who refused nasogastric tube and no cases were reported among patients
               feeding were compared.                  deemed safe on an oral diet. Four of the 5
                                                       deaths occurred in the NG group. The
                                                       difference was statistically significant
Leder et al.   1260 consecutively enrolled inpatients, There were no significant differences in
2008           630 with an NG tube in place and 630 aspiration of either liquid or puree food
USA            without at the time of assessment for consistencies dependent on presence of an NG
No Score       dysphagia. The aspiration status of all tube. The analysis was adjusted for sex, age,
               subjects was established using FEES.    or diagnostic category.
               3 trials each of both pudding and thin
               fluid consistencies were trialed.
Langdon et al. A cohort of 330 ischemic stroke         Over the study period the number of
2009           survivors were followed for 30 days to respiratory infections in tube fed and orally fed
Australia      determine whether the risk of           patients were 30/51 (59%) and 21/64 (33%),
No Score       pneumonia was higher in tube fed        respectively. The risk of pneumonia was
               patients compared with orally fed.      increased in tube fed patients (RR=4.94, 95%
                                                       CI 3.02-8.10, p<0.001).

Dziewas et al. (2004) reported an                     of aspiration, at least not in subjects
extremely high rate of pneumonia                      with mild or moderately-disabling
among 100 acute stroke patients who                   stroke. Therefore, while it remains
were fed via a nasogastric tube due to                uncertain whether NG tubes pose a
dysphagia. Most patients developed                    higher risk for the development of
pneumonia on the second or third day                  pneumonia following, a physiological
following stroke (median of 2 days,                   basis for a putative mechanism
range 0-9 days) and in some cases,                    remains unknown. Most likely, other
while the feeding tube was being used                 factors such as being bed bound,
only for gastric decompression,                       increased age and medical comorbidity
highlighting the fact that feeding tubes              confound the relationship.
are not protective from colonized oral
secretions. Factors most predictive of                Although there was no comparison
the development of pneumonia were                     group, the high incidence of
initial decreased level of consciousness              pneumonia raises troubling concerns
and facial palsy. More recently, the                  about the effectiveness of feeding
same authors (Dziewas et al. 2008)                    tubes in preventing pneumonia in
suggested that correctly-placed NG                    high-risk populations. Marum & Lim
tubes do not interfere with swallowing                (2005) also reported a higher
physiology and do not increase the risk               incidence of aspiration pneumonia and
15.   Dysphagia                                                                          pg. 38 of 60
death among geriatric patients                  meet their nutritional needs orally.
assigned to NG feeding. The results             Enteral feeding should be considered
are confounded by the fact that NG fed          after a stroke survivor has been NPO for
patients were more cognitively- and             48 hours.
functionally-impaired compared to
those on oral feeding. However, in sub          Although enteral feeding for dysphagic
group analysis the rate of pneumonia            stroke patients is a well-established
                                                practice, there is conflicting (Level 4)
was still higher among patients who
                                                evidence that nasogastric tubes reduce
accepted NG feeding compared with
                                                the risk of pneumonia.
those who refused the treatment.
In contrast to findings from these 2
studies, Nakajoh et al. (2004) reported          Enteral tube feeding should be
                                                 considered for stroke patients at risk
that the incidence of pneumonia was
                                                 of aspiration.
4.1 times greater among 73 dysphagic
stroke patients who were orally fed
                                               15.9.4 Selection of Feeding Tubes
(n=35), compared to those who
received non-oral feedings suggesting
that nasogastric tubes are protective          Enteral feeding may be required for
for pneumonia. The authors also                either brief or prolonged periods of
suggested that this protective effect          time and is used most commonly in
might be limited to patients who are           the treatment of dysphagia. As a
not bedridden.                                 result, the choice of feeding tube is
                                               dictated, in large part, by the
In contrast to these findings Landon et        anticipated length of swallowing
al. (2009) reported an increased risk in       impairment. Broadley et al. (2003)
the incidence of pneumonia associated          have identified several predictors of
with tube feeding. There was also a            prolonged dysphagia, which include
significant time-to-event effect with          initial stroke severity, dysphasia and
73% (22/30) respiratory infections in          the involvement of frontal or insular
tube-fed survivors diagnosed on days           cortex on brain imaging. However,
2-4 after stroke, and 76% (39/51) of           clinically, it can be challenging to
infections in all tube-fed survivors           accurately predict the length of time
occurring by day 7 after stroke. The           that enteral feeding will be required.
authors suggested that there may be a          Feeding tubes fall into two broad
period of increased susceptibility to          categories, nasogastric (NG) tubes,
infections in the acute post stroke            usually intended for short-term use
period. The phenomenon, “stroke-               and which are positioned directly into
induced immunodeficiency” has been             the stomach (with extensions into the
coined to describe the condition in            small bowel) or small intestine either
which there is an inhibition of cell-          percutaneously or surgically.
mediated immunity, which has been              Generally, gastro-enteric tubes are
demonstrated in animal models.                 used for long-term feeding. There are
                                               advantages and disadvantages to both
 Conclusions Regarding the Use of Non-         tube types. Nasogastric tubes have
 Oral Feeding                                  been shown to be less effective with
                                               greater side effects compared to
 There is consensus (Level 3) opinion          gastrostomy tubes for patients that
 that enteral tube feeding be used in          require a longer duration of non-oral
 stroke patients who are dysphagic and         feeding (Hull et al. 1993, Park et al.
 at high risk for aspiration or who cannot     1992), although significant mortality

15.   Dysphagia                                                               pg. 39 of 60
and morbidity has been associated                         nasal loop technique to anchor tubes
with more invasive enteric tubes, such                    securely in place, preventing
as the percutaneous endoscopic                            dislodgement and subsequent
gastrostomy (PEG) (Anderson et al.                        reinsertion. Table 15.24 presents the
2004). Anderson et al. (2004)                             results of several studies evaluating
describes the successful placement of                     the nasogastric and percutaneously
NG tubes in stroke patients using the                     placed feeding tubes.

Table 15.24 The Efficacy of Non-Oral Feeding Post Stroke: Gastric/gastro-jejunal
vs. NG Feeding Tubes
 Author/Year/                  Methods                                        Outcome
 PEDro score
Park et al.     28 days of enteral feeding using either    Treatment failure occurred in 18/19 patients in
1992            a percutaneous endoscopic                  NG group compared to 0/19 in PEG group
Scotland        gastrostomy (PEG) tube or naso-            Patients in NG group received less volume of
6 (RCT)         gastric (NG) tube was evaluated in 40      feed compared to PEG group (55% vs. 93%).
                patients (18 with stroke) with long-
                standing (>4 weeks) dysphagia.
Norton et al.   30 dysphagic stroke patients            At 6 weeks post stroke, a significantly greater
1996            randomized to receive either a          proportion of patients died in the NG group
UK              gastrostomy feeding tube (GT) or a      compared to the GT group (8 vs. 2). Patients
6 (RCT)         nasogastric tube (NG).                  in the GT group had better nutritional indices
                                                        including weight, serum albumin, mid-arm
                                                        circumference. There were no omitted feeds
                                                        among patients in the GT group compared to
                                                        at least one missed feed in 10 patients in the
                                                        NG group.
Lien et al.     8 stroke patients with a modified       24 hr esophageal acid exposure (% of time
2000            feeding tube in place allowing for both that pH readings was <4) were significantly
Taiwan          gastric (PEG) and jejunal (PEJ) feeding lower on PEJ feeding days compared to PEG
No Score        were tested for gastroesophageal        feeding days (10.1% vs. 20.6%). During meal
                reflux using 24 hr esophageal pH        infusion, 7/8 patients had less esophageal acid
                monitoring.                             exposure during PEJ feeding compared to PEG
FOOD Trial      321 acute stroke patients, from 47      Feeding with a PEG tube was associated with
2005            hospitals in 11 countries, were         an increase in the absolute risk of death or
UK              randomized to receive either a PEG      poor outcome of 7.8% (p=0.05). There was no
7 (RCT)         (n=162) or NG feeding tube (n=159)      difference in the incidence of pneumonia
                within 3 days of enrolment into the     between the groups. There were more
                study. Death and poor outcome           gastrointestinal bleeds among patients in the
                (defined as a Modified Rankin Score of NG group (18 vs. 5, p=0.005), but more
                4-5) was assessed at 6 months.          pressure sores among patients in the PEG
                                                        group (12 vs. 4, p=0.04). Only 48% of
                                                        patients allocated to treatment in the PEG
                                                        group actually received the treatment within 3
Kostadima et    41 acutely ill, ventilator dependent    At the end of weeks 2 and 3 the cumulative
al.             patients with a diagnosis of either     incidence of pneumonia was significantly
2005            stroke (n=25) or head injury (n=16)     higher in the NG compared to the gastrostomy
Greece          were randomized to receive a            group (p<0.05). At the end of the first week
6 (RCT)         gastrostomy or to NG tube for enteral the incidence of pneumonia was higher in the

15.   Dysphagia                                                                             pg. 40 of 60
                feeding. Tubes were inserted within 24     gastrostomy group although the result was not
                hours of intubation. Patients were         statistically significant.
                followed for 3 weeks and the incidence
                of pneumonia was noted and                                  Gastostomy                NG
                compared between groups. A diagnosis       Week 1            2/20                   1/20
                of pneumonia was established using         Week 2            2/18                    8/19
                previously validated criteria.             Week 3             2/16                   8/18
Beavan et al.   104 patients requiring non-oral feeding    Subjects in the nasal loop group received a
2010            following acute stroke received an NG      significantly greater volume of prescribed feeds
UK              tube which was secured using either        and fluids over 2 weeks (75% vs. 57%,
7 (RCT)         conventional means (adhesive tape)         p=0.02) and required fewer NG tubes (median
                (n=53) or a nasal loop (n = 51). The       1 vs.4). There were no differences in outcomes
                main outcome measure was the               at 3 months (death, BI scores. Death or
                proportion of prescribed feed and          dependency, length of hospital stay.
                fluids delivered via NGT in 2 weeks.
                Secondary outcomes were frequency
                of NGT insertions, treatment failure,
                tolerability, mortality; length of
                hospital stay; residential status; and
                Barthel Index at 3 months.

                                                          studies. The results are summarized in
1. Death or Poor Outcome                                  Table 15.25. While two studies
                                                          reported an increased risk of death
Three trials evaluated the risk of death                  associated with NG feeding, neither
associated with type of feeding tube                      the results from the individual studies,
(Norton et al. 1996, FOOD 2005,                           nor the pooled estimate was
Kostadima et al. 2005). The FOOD                          statistically significant. This finding
trial (2005) also assessed the risk of                    suggests that the type of feeding tube
the combined outcome of death or                          used does not increase the risk of
poor outcome (defined as a modified                       death. The results from the largest
Rankin scale score of 4-5). The                           and most important of the trials
results are difficult to pool and to                      (FOOD 2005) are presented in Figure
interpret as the patient population and                   15.4.
the length of follow-up varied between

Table 15.25 The Relative Risk of Death Associated with NG vs. Gastrostomy
Feeding Among RCTs.
       Study            Patients (n)      Length of follow-up          Relative Risk (95% CI) for Death
                                                                     ( using NG as reference condition)
Norton et al.        Acute stroke             6 weeks                       5.47 (1.16, 18.05)
1996                 (n=30)
FOOD 2005            Acute stroke             6 months                     0.98 (0.78, 1.23)
Kostadima et al.     ICU (n=41)               3 weeks                      1.43 (0.47, 4.32)
Pooled estimate                                                            1.07 (0.86, 1.33)

15.   Dysphagia                                                                              pg. 41 of 60
            Figure 15.4 NG vs. PEG Tube Feeding on Stroke Outcome (FOOD 2005)

                                                             81.0%            89.0%

  60.0%            48.0%           49.0%

  40.0%                                                              p=0.05

                           Death                             Death/Poor Outcome

                                       NG           PEG

2.Pneumonia                                          all were ventilator-dependent. Both
                                                     groups of patients were similar in
While two RCTs assess ed the                         terms of baseline characteristics and
incidence of pneumonia associated                    medical management. The authors
with feeding tube type (FOOD 2005,                   speculate that the reasons for the
Kostadima et al. 2005), unfortunately,               increase among patients with NG
the data from the FOOD trial were not                tubes may be due to: “disturbance of
reported, although the authors noted                 the pharyngoglottal refluxes that
that there was no difference in the                  prevent aspiration, dysfunction of the
proportion of patients who developed                 upper and lower oesophageal
pneumonia between groups                             sphincters and associated gastro-
(gastrostomy vs. NG). However,                       esoghageal reflux secondary to the
Kostadima et al. (2005) reported that                presence of the tube and colonization
a significantly greater proportion of                of the stomach by bacteria that may
patients fed by a NG tube developed                  subsequently migrate into the
pneumonia within 3 weeks, compared                   oropharynx and into the lower
to patients who had a gastrostomy                    respiratory tract.” It is uncertain
tube placed immediately following                    whether these results can be
admission to an ICU. The majority,                   extrapolated to a non-ventilated
but not all patients recruited for this              population.
study had suffered from a stroke and

 Conclusions Regarding Choice of                       mechanical failures compared to
 Feeding Tube                                          nasogastric feeding tubes.

 There is consensus (Level 3) opinion                  Based on the results from one large,
 that if dysphagia is severe and                       international trial, there is moderate
 expected to last more than 6 weeks, a                 (Level 1b) evidence that the type of
 gastrostomy or jejunostomy feeding                    feeding tube (nasogastric or gastro-
 tube may be indicated.                                enteric) does not affect the odds of
                                                       death or the combined outcome of
 Based on the results from two “good”                  death or poor functional outcome.
 quality RCTs, there is strong (Level 1a)
 evidence that intragastric feeding                    There is moderate (Level1b) evidence
 devices are associated with fewer                     that the risk of developing pneumonia

15.   Dysphagia                                                                       pg. 42 of 60
 is higher among ventilated patients fed
 by a naso-gastric tube compared with a                Enteral tube feeding may be necessary
 gastrostomy tube.                                     when stroke patients fail to meet their
                                                       nutritional needs orally. Gastric or
 There is moderate (Level 1b) evidence                 jejunostomy feeding tubes are preferred
 that securing naso-gastric tubes with a               over nasogastric tubes for providing
 tether-like device reduces the number                 nutrition and hydration to dysphagic
 of dislodged tubes and increases the                  patients who require non-oral support
 amount of required feed and fluids that               for more than 28 days.
 patients receive.

15.9.5 Alternative Interventions                       thermal and electrical stimulation,
In addition to texture-modified diets                  acupuncture, antihypertensive agents
and non-oral feeding, a variety of                     and decontamination of the digestive
mostly small studies have evaluated                    tract. An evaluation of a variety of
the efficacy of a variety of                           dysphagia treatment modalities is
miscellaneous treatments including                     presented in Table 15.26.

Table 15.26 Effect of Alternative Interventions in Dysphagic Stroke Patients
  Author/Year                   Methods                                     Outcome
  PEDro score
Logemann et al.   The effect of head rotation on         Head rotation did not alter the swallowing
1989              swallowing function was evaluated      efficiency of healthy subjects. In stroke
USA               on 5 lateral medullary stroke          patients head rotation improved swallowing
No Score          patients and 14 healthy adults.        “efficiency” from 21 to 50% and increased the
                                                         diameter of the upper esophageal sphincter
                                                         from 7.7 to 11.6 mm.
Rosenbek et al. In a crossover ABAB study, 7             No evidence that treatment with thermal
1991             patients received a week-long period application improved incidence of aspiration,
USA              of thermal application (chilled         penetration or residuae.
6 (RCT)          laryngeal mirror used to stroke the
                 anterior faucial pillar on both sides),
                 followed by 3 cc of water or ice
                 chips, followed by no treatment for
                 one week.
Park et al. 1997 Case reports of 4 dysphagic stroke      Improvement in swallowing function in 2/4
UK               patients receiving electrical           patients including a reduction in transit time
No Score         stimulation of the palatal area to      and absence of pooling/penetration/aspiration.
                 improve swallowing function.
Perez et al.     17 patients were randomized to          Patients in the treatment group demonstrated
1998             receive 30 mg slow release              significant improvement in mean pharyngeal
UK               nifedipine orally or placebo for 28     transit time and swallowing delay compared to
7 (RCT)          days. All patients also received        patients in the control group.
                 treatment by a speech therapist.
Arai et al. 1998 16 hypertensive stroke patients with Symptomless dysphagia improved in 10/16
Japan            symptomless dysphagia and 10            patients, as assessed by changes in serum
No Score         hypertensive patients without           substance P concentrations. No details of
                 dysphagia and 7 control patients        dosages of ACE inhibitor or length of treatment
                 were studied. To observe the            time were reported.
                 occurrence of symptomless
                 dysphagia 1 mL of Technetium Tin

15.   Dysphagia                                                                           pg. 43 of 60
                  Colloid was given during sleep by
                  nasal catheter. Hypertensive patients
                  received an ACE inhibitor.
Rosenbek et al. 45 patients were randomized to             Combing all levels of intensity, mean DST was
1998              receive treatment that included          reduced with 3 mL fluid intake at week 2 (1.17
USA               rubbing both anterior faucial pillars    sec, p=0.06). There were no significant
5 (RCT)           briskly, 3 or more times with an ice     improvements in DST using 10 mL fluid. With
                  stick. Study Group A:150 trials of       both 3 and 10 mL boluses, no intensity of
                  tactile-thermal stimulation per week;    treatment was associated with a clinically
                  Study Group B:300 trials per week;       significant improvement in P/A. score. A
                  Study Group C:450 trials per week;       statistically significant improvement in mean
                  Study Group D:600 trials per week        P/A scores was associated with 150 trials per
                  2 outcomes were assessed at 1 and        week at weeks 1 and 2 (3 mL bolus only).
                  2 weeks: Duration of Stage               Combining all levels of intensities, mean P/A
                  Transition-DST (sec) with intakes of     scores were reduced with 3 mL fluid intake at
                  3 mL and 10 mL liquid;                   weeks 1 (0.55, p=0.04 and 2 (0.59, p=0.03).
                  Penetration/Aspiration-P/A with
                  intakes of 3 mL and 10 mL liquid.
Freed et al.      Controlled trial whereby 99              Mean swallowing scores between the groups
2001              dysphagic stroke patients were           were similar at baseline. At the time of final
USA               assigned to receive either thermal-      assessment the mean swallowing scores were
No score          tactile stimulation (TS) or electrical   significantly higher among patients in the ES
                  stimulation (ES). TS was given in 3-     groups compared to the TS group (4.52 vs.
                  20 minutes daily sessions. A small       1.13). 98% of ES patients showed some
                  mirror was chilled in ice and then       improvement, whereas 27% of TS patients
                  applied to the anterior faucial arch.    remained at initial swallow score and 11% got
                  In ES treatment, the electrodes of a     worse. These results are based on similar
                  hand-held stimulator were placed on      numbers of treatments (average of 5.5 for ES
                  the neck in one of two positions until   and 6.0 for TS).
                  muscle fasciculations occurred.
                  Frequency and pulse width were
                  fixed at 80Hz and 300 ms.
                  Swallowing function was assessed
                  before and after treatment using a 0
                  (worst) to 6 (best) aspiration scale.
                  Treatment continued until patient
                  achieved a score of 5 or was
                  discharged from hospital.
Arai et al.       51 stroke patients, all with proven      Silent aspiration disappeared in 10/13 patients
2003              silent aspiration (assessed by 1 mL      given cabergoline, 10/14 patients given
Japan             of Technetium Tin Colloid was given      amantadine and 9/12 patients given imidapril
RCT (insufficient during sleep by nasal catheter) were     and in 1/12 of the control patients. Significant
data provided to studied. 39 normotensive patients         differences were noted between the collective
score-letter to   were randomized to receive               treatment group and no treatment group.
the editor)       cabergoline (0.25 mg/day) n=13,
                  amantadine (50 mg/day) n=14, or
                  no active treatment (n=12). The
                  remaining 12 patients were
                  hypertensive and received imidapril
                  (5 mg/day). Treatment lasted for 12
Crary et al.      25 patients with dysphagia               92% of the stroke patients increased functional
2004              underwent a systematic therapy           oral intake by at least one scale score following

15.   Dysphagia                                                                              pg. 44 of 60
Florida          program. Stroke patients outcomes        therapy. 48% of the stroke patients improved
No Score         were compared to 20 head/neck            1 or 2 levels, 20% improved 3 to 5 levels, and
                 cancer patients. Sessions were 50        24% improved 6 levels. Of the stroke patents
                 min per day (no weekends) with           reliant on nonoral feeding pre-therapy, 55%
                 surface electromyographic (sEMG)         progressed to total oral feeding.
                 biofeedback. Outcome measures
                 included change in functional oral
                 intake, the number of therapy
                 session to discharge, and estimated
                 cost per unit of functional charge.
Seki et al. 2005 32 stroke patients with severe           In the intervention group, the swallowing time
Japan            dysphagia were randomized to             of water and fluid food were shortened
RCT              receive acupuncture 3 x per week x       significantly. The incidence of aspiration on 3
(insufficient    4 weeks + usual care or to usual         food consistencies (water, fluid food and solid
data provided to care. Swallowing function was            food) was reduced significantly in the
score-letter to  assessed by VMBS at baseline and at      intervention group, with no corresponding
the editor)      the end of treatment.                    declines in the control group.
Gosney et al.    203 stroke patients were randomized      58/203 (29%) patients had dysphagia on
2006             to receive either selective              admission. AGNB was acquired on 2 or more
UK               decontamination of the digestive         consecutive samples in 6 patients in the
6 RCT            tract (SDD) gel or placebo 4 x daily     treatment group and 11 in the control group.
                 for 2 weeks (patients with safe          The difference between groups was not
                 swallow) or 3 weeks (unsafe              statistically significant. Four patients in the
                 swallow). Outcomes included the          treatment group acquired either pneumonia or
                 incidence of colonization of aerobic     septicemia during the study period compared
                 gram negative bacilli (AGNB) and         to 10 in the placebo group (p=0.029).
                 septicemia or respiratory tract          Mortality between groups was similar (9 in
                 infections during hospital stay.         treatment group vs. 11 in control group).
Ebihara et al.   105 chronic stroke patients residing     Nasal inhalation of BPO shortened LTSR,
2006             in nursing homes were randomized         compared with that of lavender oil and distilled
Japan            to either a study group, black pepper    water (P < .030. The number of swallowing
5 (RCT)          oil (BPO)(n=35), or one of 2 control     movements for 1 minute during the nasal
                 groups, lavender oil (LO) (n=35), or     inhalation of BPO increased (P < .001).
                 distilled water (n=35). Nasal
                 inhalation of 100цL of one of the 3
                 substances was administered to the
                 nostrils with a paper stick. Subjects
                 received the treatment 1 minute
                 before each meal during the 30-day
                 study period. Latency of the
                 swallowing reflex (LTSR) and the
                 number of swallowing movements
                 were assessed.
Power et al.     16 dysphagic stroke subjects were        Compared with baseline, no change was
2006             randomized to receive treatment          observed in the speed of laryngeal elevation,
UK               consisting of stimulation of the         pharyngeal transit time, or aspiration severity
4 (RCT)          anterior faucial pillar with either no   within subjects or between groups for either
                 (sham) stimulation or stimulation at     active or sham stimulation.
                 a frequency of 0.2 Hz for 10 minutes
                 (5 on each side). Swallowing was
                 assessed before and 60 min after
                 electrical or sham stimulation.
                 Swallowing measures included

15.   Dysphagia                                                                             pg. 45 of 60
                  laryngeal closure (initiation and
                  duration) and pharyngeal transit
                  time, taken from VMBS study.
                  Aspiration severity was assessed
                  using an 8-point scale.
Robbins et al.    10 stroke patients (6 acute and 4        There was a significant increase in maximum
2007              chronic) performed an 8-week             isometric pressure (anterior and posterior
USA               isometric lingual exercise program       tongue) over the study period. There was
No Score          by compressing an air-filled bulb        significant reduction in overall residue for 3
                  between the tongue and the hard          bolus conditions (3 mL effortful swallow, 10 mL
                  palate. This exercise was repeated       liquid and 3 mL liquid). There were significant
                  10 x, 3 times a day, 3 days a week.      improvements 3/11 subscales of the QoL
                  Isometric and swallowing lingual         questionnaire.
                  pressures, bolus flow parameters,
                  diet, and a dysphagia-specific quality
                  of life questionnaire (QoL)were
                  assessed at baseline, week 4, and
                  week 8.
Shimizu et al.    The pharyngeal transit time (PTT) of     The PTT of control subjects was unchanged
2008              10 elderly subjects with stroke was      from first to second assessment. (2.4 vs. 2.6
Japan             assessed using VMBS before and           sec). The PTT of subjects receiving the ACE
No Score          after 6 weeks of therapy with 5 mg       inhibitor significantly decreased from baseline
                  imidapril. The results were compared     to 5 weeks (2.5 vs. 1.6 sec, p<0.01).
                  with 10 age-matched healthy              Abnormalities in the oral and esophageal
                  subjects.                                phases were not altered by treatment.
Bϋlow et al.      25 stroke patients from 3 European       While subjects in both groups improved over
2008              swallowing centers were randomized       the treatment period there were no statistically
Sweden            to receive a 3-week trial (15            significant differences on any of the outcomes.
3 (RCT)           sessions) of either neuromuscular
                  electrical stimulation (NMES) or
                  traditional swallowing therapy (TT).
                  Measurements including
                  videoradiographic swallowing
                  evaluation, nutritional status, oral
                  motor function test, and a visual
                  analog scale (VAS) for self-
                  evaluation of complaints, were
                  assessed before and after treatment.
Khedr et al.      26 patients with post-stroke             At baseline the mean dysphgaia score for the
2009              dysphagia due to single hemispheric      control group was 3.7 vs. 3.4 for the real rTMS
Egypt             stroke were randomly allocated to        group. By 2 months the real rTMS groups’
6 (RCT)           receive real (n = 14) or sham (n =       mean score was approximately 1.0 vs. 3.0 for
                  12) rTMS of the affected motor           the control group. There was a significant time
                  cortex. Each patient received a total    x group interaction.
                  of 300 rTMS pulses at an intensity of
                  120% hand motor threshold for five
                  consecutive days. Clinical ratings of
                  dysphagia were assessed using the
                  Dysphagic Outcome and Severity
                  Scale before and after the last
                  session and then again after 1 and 2
                  months. Scores ranged from 1 (no
                  dysphagia) to IV (obvious dysphagia

15.   Dysphagia                                                                              pg. 46 of 60
                  precluding oral feeding). All subjects
                  received standard medical and
                  physical therapy.
Permsirivanich    23 stroke patients with dysphagia        Before therapy, 73% of the RST group and
et al. 2009       persisting for > 2 weeks were            83% of the NMES group required non-oral
Thailand          randomized to receive either             feeding (FOIS levels 1-3). At the end of the
6 (RCT)           rehabilitation swallowing therapy        study period, 75% of the RST group and 90%
                  (RST) or neuromuscular electrical        of the NMES group could manage oral intake
                  stimulation therapy (NMES). The          (FOIS groups 4-7). The differences in
                  subjects received 60 minutes of          proportions were not statistically significant.
                  either RST or NMES treatment for         18% of RST and 17% of the NMES subjects
                  five consecutive days, had two days      had attained a FOIS score of 7. There was a
                  off and then five more consecutive       significant difference in the change scores,
                  days of treatment for a four-week        favouring the NMES group (+3.17 vs. 2.46,
                  period or until they reached             p<0.001)
                  functional oral intake scale (FOIS)
                  level 7. FOIS, the primary outcome
                  measure, was assessed before /after
                  treatment. FOIS score 1=NPO, FOIS
                  score 7= oral diet, no restrictions
Lim et al. 2009   36 received thermal-tactile              28 persons completed the study. Median
Korea             stimulation (TTS) treatment only         swallowing scores for the control group, on
No Score          (control group) or TTS +                 semi-solids consistency, before and after
                  neuromuscular electrical stimulation,    treatment were 3 and 4, and 2 and 4 for the
                  applied simultaneously. Swallowing       experimental group. Median PAS scores for the
                  function was assessed before and 4       control group before and after treatment were
                  weeks after treatment using the          3.5 and 4 (indicating a worsening) and 5.5 and
                  swallow function scoring system          2.5 for the experimental group. The differences
                  (scoring: 0-6 with lower scores          in both scores between the two groups were
                  indicating greater severity), the        significant. There was significantly greater
                  penetration-aspiration scale (PAS)       improvement in pharyngeal transit time in the
                  (scoring: 1-8 with higher scores         experimental group.
                  indicating increasing
                  aspiration/penetration) and
                  pharyngeal transit time. Semi-solid
                  and liquid consistencies were
Khedr & Abo-      22 patients with acute ischemic          Among patients with lateral medullary there
Elfetoh 2010      stroke with lateral medullary            were significant improvements in dysphagia
Egypt             syndrome or brainstem infarction         scores and BI scores in the active rTMS group
6 (RCT)           severe bulbar manifestation were         compared with the sham group that were
                  randomly allocated to receive active     maintained over the study period. Among
                  (n=11) or sham (n=11) repetitive         patients with other types of brainstem infarcts
                  transcranial magnetic stimulation        who received rTMS, there was significant
                  (rTMS) of the oesophageal motor          improvement in dysphagia scale scores
                  cortex. Each patient received 300        compared with sham treatment.
                  rTMS pulses at 3 Hz and an intensity
                  of 130% resting motor threshold to
                  each hemisphere for five consecutive
                  days. Outcomes were assessed
                  before and immediately after the last
                  session, and then again after 1 and 2
                  months using a 4-point dysphagia
                  grading scale, NIHSS, BI and the

15.   Dysphagia                                                                              pg. 47 of 60
                  Hemiplegic Stroke Scale.
Jayasekeran et    50 acute dysphagic stroke patients      Patients who received the active form of PES
al. 2010          were assigned randomly to receive       experienced significantly fewer episodes of
UK                either active or sham pharyngeal        aspiration, greater improvement in DSR and
8 (RCT)           electrical stimulation (PES) once       remained in hospital for a shorter period of
                  daily for 3 days. (n = 28). The         time compared with patients who received
                  primary end point was the reduction     sham treatment.
                  of airway aspiration at 2 weeks
                  postintervention assessed using VFS.
                  Additional outcomes included scores
                  on a Dysphagia Severity Rating
                  (DSR) rating scale

Discussion                                               ACE inhibitor, imidapril on a small
                                                         group of elderly stroke subjects. The
The use of medications to improve                        use of an ACE inhibitor was associated
swallowing function was assessed in                      with a reduction in pharyngeal transit
three small RCTs (Perez et al. 1998,                     times. ACE inhibitors are believed to
Kobayashi et al. 1996, Arai et al.                       confer benefit through an increase in
2003). Perez et al. reported that                        serum substance P concentration. All
Nifidipine, a calcium channel blocker                    of the studies evaluating the use of
improved pharyngeal transit time and                     drugs to improve parameters of
swallow delay. The mechanism                             swallowing function were small.
through which the benefits were
achieved is uncertain although the                       Two studies authored by Rosenbeck et
authors speculated that it might be                      al. (1991, 1998) evaluated the
mediated by the reduction of                             effectiveness of cold stimulus to
esophageal spasm, through the action                     improve specific aspects of the
of dihydropyridines on the calcium                       swallowing mechanism. The earlier
channels, of nonvascular smooth                          study evaluating 7 patients using a
muscle. Kobayashi et al. (1996)                          crossover design, failed to
reported improved latency of response                    demonstrate a significant benefit of
in a small crossover designed study                      the treatment. A second, slightly
evaluating levadopa. Arai et al. (2003)                  larger study, evaluated the effect of
evaluated the effectiveness of the                       four increasing intensities of tactile
dopamine agonists, cabergoline,                          thermal application. The study design
reported to have fewer side effects                      did not include a control group. The
than levadopa, and the drug                              results did not favour a single level of
Amantadine, an antidyskinetic and                        intensity of treatment. In the absence
reported a significant reduction in the                  of a control group, conclusions
incidence of silent aspiration among                     regarding the effectiveness of these
normotensive patients with stroke,                       treatments could not be drawn.
compared to a no drug control
condition. Among patients with known                     The use of antimicrobial agents to
hypertension, the angiotensin-                           reduce the bacterial load in the
converting enzyme (ACE) imidaril                         digestive tract of stroke patients, in an
hypochloride also reduced the                            effort to reduce the incidence of
incidence of silent aspiration                           aspiration pneumonia was assessed in
compared to control. Shimizu et al.                      a single RCT (Gosney et al. 2006).
(2008) also evaluated the effect of the                  Treatment with the antimicrobial gel

15.   Dysphagia                                                                           pg. 48 of 60
removed the presence of a variety of                    established when compared to
aerobic gram-negative bacilli, which                    traditional swallowing therapy.
resulted in a decreased incidence of
septicemia and pneumonia.                               Ebihara et al. (2007) provided
                                                        preliminary evidence that inhalation of
Although electrical stimulation is                      black pepper oil can help to prevent
widely used clinically in the United                    the development of aspiration
States, there is a lack of evidence                     pneumonia. The authors speculated
supporting its use. A recent meta-                      that the insular cortex plays a role in
analysis (Carnaby –Mann & Crary                         both dysphagia and appetite
2007), which included the results from                  stimulation, and since black pepper oil
7 trials reported a large effect size                   is an appetite stimulant and increases
associated with the treatment. The                      blood flow to this area, it might be an
participants in the individual trials                   effective approach to the treatment of
were dysphagic due to a variety of                      dysphagia.
conditions, including stroke. The
results from the 3 trials included in                   The results of studies using an RCT to
the present review suggested that the                   evaluate an alternative intervention
effectiveness of the electrical                         are presented in table 15.27.
stimulation has not yet been

Table 15.27 Summary of RCTs Evaluating an Alternative Interventions
Intervention              Author/Year/(PEDro)                     n                 Result
Electrical stimulation    Power et al. 2006 (4)                  16                   -
                          Bϋlow et al. 2008 (3)                  25                   -
                          Permsirivanich et al. 2009             23                  -/+
                          Jayasekeran et al. 2010 (8)            31                   +
Thermal stimulation       Rosenbek et al. 1991(6)                7                    -
                          Rosenbek et al. 1998 (5)              45                    -
Nifedipine                Perez et al. 1998 (7)                 17                    +
SDD (selective            Gosney et al. 2006 (6)                203                   +
decontamination of the
digestive tract)
Black pepper oil           Ebihara et al. 2006 (5)              105                   +
Transcranial magnetic      Khedr et al 2009 (6)                 26                    +
stimulation                Khedr & Abo-Elfetoh                  22                    +
+ evidence of benefit when compared to control condition
- no evidence of benefit when compared to control condition
                                                         be used to improve specific aspects of
 Conclusions Regarding Alternative                       swallowing following stroke.
 Interventions in Dysphagia
                                                         There is moderate (Level 1b) evidence
 There is strong (Level 1a) evidence that                that selective decontamination of the
 transcranial magnetic stimulation                       digestive tract can help to reduce the
 improves swallowing function post                       incidence of pneumonia.
                                                         There is strong (Level 1a) evidence that
 There is moderate (Level 1b) evidence                   thermal stimulation does not improve
 that Nifedipine and black pepper oil can                swallowing function post stroke.

15.   Dysphagia                                                                       pg. 49 of 60
                                                EMG treatment can be used to improve
There is conflicting (Level 4) evidence         swallowing function post stroke.
that electrical stimulation can improve
swallowing function post stroke.                 A variety of alternative treatments can
                                                 be used to improve swallowing
There is limited (Level 2) evidence that         function post stroke.
head rotation, lingual exercises and

15.   Dysphagia                                                              pg. 50 of 60
15.10 Summary
1. The incidence of dysphagia appears            feeding assistance or supervision to
  to be quite high following acute               all stroke survivors.
  stroke, with between one-third to two-
  thirds of all stroke patients affected.      10. There is consensus (Level 3)
                                                 opinion that a dietician should assess
2. VMBS studies are the “gold                    the nutrition and hydration status of
  standard” for diagnosing dysphagia             all stroke patients who fail a
  and aspiration.                                swallowing screening.

3. The incidence of aspiration in the          11. There is consensus (Level 3)
  acute phase of stroke varies from 21-          opinion that dysphagic stroke
  42% and decreases to less than 12%             patients typically require diets with
  by 3 months post stroke. Between               modified food and liquid textures.
  one-third and one-half of patients
  who aspirate following stroke are            12. For patients who require assistance
  silent aspirators.                             to feed, there is a consensus (Level 3)
                                                 opinion that low-risk feeding
4. Aspiration appears to be associated           strategies by trained personnel
  with an increase in the incidence of           should be employed.
  pneumonia. The risk of developing
  pneumonia appears to be                      13. There is consensus (Level 3)
  proportional to the severity of                opinion that for patients on modified
  aspiration.                                    diets that a dietitian should be
                                                 consulted to ensure that the modified
5. There is limited (Level 2) evidence           diet is nutritionally adequate and
  that dysphagia screening protocols             appropriate, and to consult the stroke
  can reduce the incidence of                    survivor or substitute decision-
  pneumonia.                                     maker, to ensure that the modified
                                                 diet is as appealing as possible.
6. There is consensus (Level 3) opinion
  that acute stroke survivors should be        14. There is limited (Level 2) evidence
  NPO until swallowing ability has been          that dysphagia diets reduce the
  determined.                                    incidence of aspiration pneumonia.
                                                 There is moderate (Level 1b)
7. There is consensus (Level 3) opinion          evidence that thickened fluids result
  that a trained assessor should screen          in fewer episodes of aspiration and
  all acute stroke survivors for                 penetration compared with thin fluids
  swallowing difficulties.                       among dysphagic individuals
                                                 following stroke.
8. There is consensus (Level 3) opinion
  that a speech and language                   15. There is limited (Level 2) evidence
  pathologist should assess all stroke           that patients requiring texture-
  survivors who fail swallowing                  modified diets including thickened
  screening and identify the                     fluids can safely consume thin fluids
  appropriate course of treatment.               between meals without increasing
                                                 their risk of pneumonia.
9. There is consensus (Level 3) opinion
  that an individual trained in low-risk       16. There is moderate (Level 1b)
  feeding strategies should provide              evidence that Nifedipine improves

15.   Dysphagia                                                              pg. 51 of 60
  specific aspects of swallowing                  feeding tubes.
  function following stroke.
                                                24. Although enteral feeding for
17. There is limited (Level 2) evidence           dysphagic stroke patients is a well-
  that head rotation can improve                  established practice, there is only
  swallowing function in lateral                  moderate (Level 1b) evidence that its
  medullary stroke patients.                      use reduces the risk of pneumonia.
                                                  There is conflicting (Level 4) evidence
18. There is moderate (Level 1b)                  that nasogastric tubes reduce the risk
  evidence that thermal stimulation               of pneumonia. There is moderate
  does not improve swallowing                     (Level 1b) evidence that the risk of
  mechanics post stroke.                          developing pneumonia is higher
                                                  among ventilated patients fed by a
19. There is moderate (Level 1b)                  naso-gastric tube compared with a
  evidence that a short course of                 gastrostomy tube.
  formal dysphagia therapy does not
  alter clinical outcomes. Based on the         25. Based on the results from one
  result from a single RCT, there is              large, international trial, there is
  moderate (Level 1b) evidence that a             moderate (Level 1b) evidence that the
  one-month dysphagia intervention                type of feeding tube (nasogastric or
  program does not improve the                    gastro-enteric) does not affect the
  likelihood of returning to a normal             odds of death or the combined
  diet by six months. However, there is           outcome of death or poor functional
  also moderate (Level 1b) evidence               outcome.
  that such a program may reduce the
  likelihood of chest infections and            26. There is moderate (Level 1b)
  death or institutionalization.                  evidence that securing naso-gastric
                                                  tubes with a tether-like device
20. There is consensus (Level 3)                  reduces the number of dislodged
  opinion that enteral tube feeding be            tubes and increases the amount of
  used in stroke patients at high risk of         required feed and fluids that patients
  aspiration or for those who cannot              receive.
  meet their nutritional need orally.
  Enteral feeding should be considered          27. There is moderate (Level 1b)
  after a stroke survivor has been NPO            evidence that Nifedipine and black
  for 48 hours.                                   pepper oil can be used to improve
                                                  specific aspects of swallowing
21. There is consensus (Level 3)                  following stroke.
  opinion that if dysphagia is severe
  and expected to last more than 6              28. There is strong (Level 1a) evidence
  weeks, a gastrostomy or jejunostomy             that transcranial magnetic stimulation
  feeding tube may be indicated.                  improves swallowing function post
22. There is limited (Level 2) evidence
  that enteral feeding tube can deliver         29. There is moderate (Level 1b)
  adequate nutrition and hydration to             evidence that selective
  stroke survivors.                               decontamination of the digestive tract
                                                  can help to reduce the incidence of
23. There is strong (Level 1a) evidence           pneumonia.
  that intragastric feeding devices are
  associated with fewer mechanical              30. There is strong (Level 1a) evidence
  failures compared to nasogastric                that thermal stimulation does not

15.   Dysphagia                                                              pg. 52 of 60
  improve swallowing function post               evidence that electrical stimulation
  stroke.                                        can improve swallowing function post
31. There is conflicting (Level 4)

15.   Dysphagia                                                           pg. 53 of 60
Diagnosis and treatment of swallowing                 Bach DB, Pouget S, Belle K, Kilfoil M, Alfieri M,
  disorders (dysphagia) in acute-care stroke            McEvoy J, Jackson G. An integrated team
  patients (Evidence report/technology                  approach to the management of patients. J
  assessment No. 8). AHCPR Publication                  Allied Health 1989; p. 459-468.
  No.99-E024 . 1999. Agency for Health Care
  Policy and Research, Rockville,MD.                  Barer DH. The natural history and functional
                                                        consequences of dysphagia after
Allison MC, Morris AJ, Park RH, Mills PR.               hemispheric stroke. J Neurol Neurosurg
   Percutaneous endoscopic gastrostomy tube             Psychiatry 1989;52:236-241.
   feeding may improve outcome of late
   rehabilitation following stroke. J R Soc Med       Beavan J, Conroy SP, Harwood R, Gladman JR,
   1992;85:147-149.                                     Leonardi-Bee J, Sach T, Bowling T, Sunman
                                                        W, Gaynor C. Does looped nasogastric tube
Alzate GD, Coons H, Elliott J, Cary PH.                 feeding improve nutritional delivery for
   Percutaneous gastrostomy for jejunal                 patients with dysphagia after acute stroke?
   feeding: a new technique. AJR                        A randomised controlled trial. Age Ageing
   1986;147:822-825.                                    2010;39:624-630.

Anderson MR, O'Connor M, Mayer P, O'Mahony            Bounds JV, Wiebers DO, Whisnant JP, Okazaki
  D, Woodward J, Kane K. The nasal loop                 H. Mechanism and timing of deaths from
  provides an alternative to percutaneous               cerebral infarctions. Stroke 1981;12(4):
  endoscopic gastrostomy in high-risk                   474-477.
  dysphagic stroke patients. Clin Nutr
  2004;23:501-506.                                    Broadley S, Croser D, Cottrell J, et al.
                                                        Predictors of prolonged dysphagia following
Antonios N, Carnaby-Mann G, Crary M, Miller L,          acute stroke. J Clin Neurosci 2003;10:300-
  Hubbard H, Hood K, Sambandam R, Xavier                305.
  A, Silliman S. Analysis of a physician tool for
  evaluating dysphagia on an inpatient stroke         Brown M, Glassenberg M. Mortality factors in
  unit: the modified Mann Assessment of                 patients with acute stroke. JAMA
  Swallowing Ability. J Stroke Cerebrovasc Dis          1973;224:1493-1495.
                                                      Bulow M, Speyer R, Baijens L, Woisard V,
Arai T, Yasuda Y, Takaya T, et al. ACE                  Ekberg O. Neuromuscular electrical
  inhibitors and symptomless dysphagia.                 stimulation (NMES) in stroke patients with
  Lancet 1998;352:115-116.                              oral and pharyngeal dysfunction. Dysphagia
Arai T, Sekizawa K, Yoshimi N, Toshima S,
  Fujiwara H. Cabergoline and silent aspiration       Carnaby G, Hankey GJ, Pizzi J. Behavioural
  in elderly patients with stroke. J Am Geriatr         intervention for dysphagia in acute stroke: a
  Soc 2003;51:1815-1816.                                randomised controlled trial. Lancet Neurol
                                                        2006; 5(1):31-37.
Arms R, Dines D, Tinstman T. Aspiration
  pneumonia. Chest 1974;65:136-139.                   Carnaby-Mann GD, Crary MA. Examining the
                                                        evidence on neuromuscular electrical
Aviv JE. Prospective, randomized outcome                stimulation for swallowing: a meta-analysis.
  study of endoscopy versus modified barium             Arch Otolaryngol Head Neck Surg
  swallow in patients with dysphagia.                   2007;133:564-571.
  Laryngoscope 2000; 110(4):563-574.
                                                      Chen YM, Ott DJ, Peele VN, Gelfand DW.
Aviv JE, Kaplan ST, Thomson JE, Spitzer J,              Oropharynx in patients with cerebrovascular
  Diamond B, Close LG. The safety of flexible           disease: evaluation with videofluoroscopy.
  endoscopic evaluation of swallowing with              Radiology 1990;176:641-643.
  sensory testing (FEESST): an analysis of
  500 consecutive evaluations. Dysphagia              Chong MS, Lieu PK, Sitoh YY, Meng YY, Leow
  2000; 15(1):39-44.                                    LP. Bedside clinical methods useful as
                                                        screening test for aspiration in elderly

15.   Dysphagia                                                                           pg. 54 of 60
  patients with recent and previous strokes.            thick consistency in stroke patients. Nutr
  Ann Acad Med Singapore 2003;32:790-794.               Clin Pract 2009;24:414-418.

Churchill M, Grimm S, Reding M. Risks of             Dziewas R, Ritter M, Schilling M, et al.
  diuretic usage following stroke. Neurorehabil        Pneumonia in acute stroke patients fed by
  Neural Repair 2004;18:161-165.                       nasogastric tube. J Neurol Neurosurg
                                                       Psychiatry 2004;75:852-856.
Clayton J, Jack CI, Ryall C, Tran J, Hilal E,
   Gosney M. Tracheal pH monitoring and              Dziewas R, Warnecke T, Hamacher C, et al. Do
   aspiration in acute stroke. Age Ageing 2006;        nasogastric tubes worsen dysphagia in
   35(1):47-53.                                        patients with acute stroke? BMC Neurol
Collins MJ, Bakheit AM. Does pulse oximetry
  reliably detect aspiration in dysphagic stroke     Ebihara T, Ebihara S, Maruyama M et al. A
  patients? Stroke 1997; 28(9):1773-1775.              randomized trial of olfactory stimulation
                                                       using black pepper oil in older people with
Crary MA, Carnarby GD, Groher ME, Helseth E.           swallowing dysfunction. J Am Geriatr Soc.
  Functional Benefits of Dysphagia Therapy             2006;54:1401-1406.
  Using Adjunctive sEMG Biofeedback.
  Dysphagia 2004;19:160-164.                         Edmiaston J, Connor LT, Loehr L, Nassief A.
                                                       Validation of a Dysphagia screening tool in
Daniels SK, McAdam CP, Brailey K, Foundas              acute stroke patients. Am J Crit Care
  AL. Clinical assessment of swallowing and            2010;19:357-364.
  prediction of dysphagia severity. American
  Journal of Speech-Language Pathology               Falsetti P, Acciai C, Palilla R, Bosi M, Carpinteri
  1997;6:17-24.                                         F, Zingarelli A, Pedace C, Lenzi L.
                                                        Oropharyngeal dysphagia after stroke:
Daniels SK, Brailey K, Priestly DH, Herrington          incidence, diagnosis, and clinical predictors
  LR, Weisberg LA, Foundas AL. Aspiration in            in patients admitted to a neurorehabilitation
  patients with acute stroke. Arch Phys Med             unit. J Stroke Cerebrovasc Dis 2009;18:329-
  Rehabil 1998;79:14-19.                                335.

Dennis M. The FOOD (Feed or Ordinary Diet)           Finegold SM. Aspiration pneumonia. Reviews
  Trial. 1997 North American Stroke Meeting,            of Infectious Diseases 1991;13 (Suppl
  Montreal, Quebec, October 18, 1997.                   9):S737-742.

Dennis MS, Lewis SC, Warlow C. Effect of             Finestone HM, Greene-Finestone LS, Wilson
  timing and method of enteral tube feeding             ES, Teasell RW. Malnutrition in stroke
  for dysphagic stroke patients (FOOD): a               patients on the rehabilitation service at
  multicentre randomised controlled trial.              follow-up: prevalence and predictors. Arch
  Lancet 2005;365:764-772.                              Phys Med Rehabil 1995;76:310-316.

DePippo KL, Holas MA, Reding MJ, Mandel FS,          Finestone HM, Greene-Finestone LS, Wilson
  Lesser ML. Dysphagia therapy following                ES, Teasell RW. Prolonged length of stay
  stroke: A controlled trial. Neurology                 and reduced functional improvement rate in
  1994;44:1655-1660.                                    malnourished stroke rehabilitation patients.
                                                        Arch Phys Med Rehabil 1996;77:340-345.
DePippo KL, Holas MA, Reding MJ. The Burke
  dysphagia screening test: validation of its        Finestone HM, Fisher J, Greene-Finestone LS,
  use in patients with stroke. Arch Phys Med            Teasell RW, Craig ID. Sudden death in the
  Rehabil 1994;75:1284-1286                             dysphagic stroke patient--a case of airway
                                                        obstruction caused by a food bolus: a brief
DePippo KL, Holas MA, Reding MJ. Validation of          report. Am J Phys Med Rehabil
  the 3-oz water swallow test for aspiration            1998;77:550-552.
  following stroke. Arch Neurol 1992;49:1259-
  1261.                                              Finestone HM, Foley NC, Woodbury MG,
                                                        Greene-Finestone L. Quantifying fluid intake
Diniz PB, Vanin G, Xavier R, Parente MA.                in dysphagic stroke patients: A preliminary
  Reduced incidence of aspiration with spoon-           comparison of oral and nonoral strategies.
                                                        Arch Phys Med Rehabil 2001;82:1744-1746.

15.   Dysphagia                                                                           pg. 55 of 60
Foley N, Teasell R, Salter K, Kruger E, Martino           screening protocols prevent pneumonia.
   R. Dysphagia treatment post stroke: a                  Stroke 2005; 36(9):1972-1976.
   systematic review of randomised controlled
   trials. Age Ageing 2008;37:258-264                  Hinds NP, Wiles CM. Assessment of swallowing
                                                         and referral to speech and language
Freed ML, Freed L, Chatburn RL, Christian M.             therapists in acute stroke. QJM
   Electrical stimulation for swallowing                 1998;91:829-835
   disorders caused by stroke. Respir Care
   2001; 46(5):466-474.                                Hinsdale JC, Lipkowitz GS, Pollock TW, Hoover
                                                         EL, Jaffe BM. Prolonged enteral nutrition in
Garon BR, Engle M, Ormiston C. Reliability of            malnourished patients with nonelemental
  the 3 oz water swallow test utilizing cough            feeding: reappraisal of surgical technique,
  reflex as sole indicator of aspiration. J Neuro        safety and costs. Am J Surg 1985;149:334-
  Rehab 1995;9:143.                                      338.

Garon BR, Engle M, Ormiston C. A randomized            Holas MA, DePippo KL, Reding MJ. Aspiration
  control trial to determine the effects of              and relative risk of medical complications
  unlimited oral intake of water in patients             following stroke. Arch Neurol
  with identified aspiration. J Neurol Rehabil           1994;51:1051-1053.
                                                       Horner J, Massey EW. Silent aspiration
Gordon C, Hewer RL, Wade DT. Dysphagia in                following stroke. Neurology 1988;38:317-
  acute stroke. Br Med J 1987;295:411-414.               319(a).

Gosney M, Martin MV, Wright AE. The role of            Horner J, Massey EW, Riski JE, Lathrop DL,
  selective decontamination of the digestive             Chase KN. Aspiration following stroke:
  tract in acute stroke. Age Ageing 2006;                Clinical correlates and outcome. Neurology
  35(1):42-47.                                           1988;38:1359-1362(b).

Gottlieb D, Kipnis M, Sister E, Vardi Y, Brill S.      Hull MA, Rawlings J, Murray FE, Field J,
  Validation of the 50 ml3 drinking test for             McIntyre AS, Mahida YR, Hawkey CJ, Allison
  evaluation of post-stroke dysphagia. Disabil           SP. Audit of outcome of long-enteral
  Rehabil 1996; 18(10):529-532.                          nutrition by percutaneous endoscopic
                                                         gastrostomy. Lancet 1993;341:869-872.
Goulding R, Bakheit AM. Evaluation of the
  benefits of monitoring fluid thickness in the        Huxley EJ, Viroslav J, Gray WR, Pierce AK.
  dietary management of dysphagic stroke                 Pharyngeal aspiration in normal adults and
  patients. Clin Rehabil 2000;14:119-124.                patients with depressed consciousness. Am
                                                         J Med 1978;64:564-568.
Groher ME. Bolus management and aspiration
  pneumonia in patients with pseudobulbar              James A, Kapur K, Hawthorne AB. Long-term
  dysphagia. Dysphagia 1987;1(4):215-216.                outcome of percutaneous endoscopic
                                                         gastrostomy feeding in patients with
Gustke RF, Varma RR, Soergel KH. Gastric                 dysphagic stroke. Age Ageing
  reflux during profusion of the proximal small          1998;27(6):671-676.
  bowel. Gastroenterology 1970;59:890-895.
                                                       Jayasekeran V, Singh S, Tyrrell P, Michou E,
Hanning C, Wuttge-Hanning A, Hormann M,                  Jefferson S, Mistry S, Gamble E, Rothwell J,
  Hermann I. A cinematographic study of the              Thompson D, Hamdy S. Adjunctive
  pathologic mechanism of aspiration                     functional pharyngeal electrical stimulation
  pneumonia. Fortschv Rontgenstr                         reverses swallowing disability after brain
  1989;159(3):260-267.                                   lesions. Gastroenterology 2010;138:1737-
Heart and Stroke Foundation of Ontario.
  Improving Recognition and Management of              Jean A. Brain stem control of swallowing:
  Dysphagia in Acute Stroke. 2002.                       neuronal network and cellular mechanisms.
                                                         Physiol Rev 2001;81:929-969
Hinchey JA, Shephard T, Furie K, Smith D,
  Wang D, Tonn S. Formal dysphagia

15.   Dysphagia                                                                          pg. 56 of 60
Johnson ER, McKenzie SW, Sievers A.                    compared with fiberoptic endoscopic
  Aspiration pneumonia in stroke. Arch Phys            examination of swallowing (FEES) in
  Med Rehabil 1993;74:973-976.                         determining the risk of aspiration in acute
                                                       stroke patients. Dysphagia 2001; 16(1):1-6.
Khedr EM, bo-Elfetoh N, Rothwell JC.
  Treatment of post-stroke dysphagia with           Lim KB, Lee HJ, Lim SS, Choi YI.
  repetitive transcranial magnetic stimulation.        Neuromuscular electrical and thermal-tactile
  Acta Neurol Scand 2009;119:155-161.                  stimulation for dysphagia caused by stroke:
                                                       a randomized controlled trial. J Rehabil Med
Khedr EM, bo-Elfetoh N. Therapeutic role of            2009;41:174-178.
  rTMS on recovery of dysphagia in patients
  with lateral medullary syndrome and               Lin LC, Wang SC, Chen SH, Wang TG, Chen
  brainstem infarction. J Neurol Neurosurg             MY, Wu SC. Efficacy of swallowing training
  Psychiatry 2010;81:495-499.                          for residents following stroke. J Adv Nurs
Kidd D, Lawson J, Nesbitt R, MacMahon J. The
   natural history and clinical consequences of     Linden P, Siebens AA. Dysphagia: predicting
   aspiration in acute stroke. Quarterly J Med         laryngeal penetration. Arch Phys Med
   1995;88:409-413.                                    Rehabil 1983;64:281-284.

Kobayashi H, Nakagawa T, Sekizawa K, Arai H,        Logemann JA. Evaluation and treatment of
  Sasaki H. Levodopa and swallowing reflex.           swallowing disorders. San Diego, CA:
  Lancet 1996; 348(9037):1320-1321.                   College-Hill Press, 1983.

Kostadima E, Kaditis AG, Alexopoulos EI,            Logemann JA, Kahrilas PJ, Kobara M, Vakil NB.
  Zakynthinos E, Sfyras D. Early gastrostomy          The benefit of head rotation on
  reduces the rate of ventilator-associated           pharyngoesophageal dysphagia. Arch Phys
  pneumonia in stroke or head injury patients.        Med Rehabil 1989;70:767-771.
  Eur Respir J 2005; 26(1):106-111.
                                                    Logemann JA, Veis S, Colangelo L. A screening
Langdon PC, Lee AH, Binns CW. High incidence          procedure for oropharyngeal dysphagia.
  of respiratory infections in 'nil by mouth'         Dysphagia 1999;14:44-51
  tube-fed acute ischemic stroke patients.
  Neuroepidemiol 2009;32:107-113.                   Lucas CE, Yu P, Vlahos A, Ledgerwood AM.
                                                      Lower esophageal sphincter dysfunction
Langmore SE, Terpenning MS, Schork A, Chen            often precludes safe gastric feeding in
  Y, Murray JT, Lopatin D, Loesche WJ.                stroke patients. Arch Surg 1999;134:55-
  Predictors of aspiration pneumonia: how             58.
  important is dysphagia? Dysphagia 1998
  Spring;13(2):69-81.                               Mann G, Hankey GJ, Cameron D. Swallowing
                                                      function after stroke: prognosis and
Leder SB, Espinosa JF. Aspiration risk after          prognostic factors at 6 months. Stroke
  acute stroke: comparison of clinical                1999;30:744-748.
  examination and fiberoptic endoscopic
  evaluation of swallowing. Dysphagia 2002;         Mann G. The Mann Assessment of Swallowing
  17(3):214-218.                                      Ability: MASA. Philadelphia:Delmar
                                                      Thompson Learning. 2002.
Leder SB, Suiter DM. Effect of nasogastric
  tubes on incidence of aspiration. Arch Phys       Mamun K, Lim J. Role of nasogastric tube in
  Med Rehabil 2008;89:648-651                         preventing aspiration pneumonia in patients
                                                      with dysphagia. Singapore Med J
Lien HC, Chang CS, Chen GH. Can                       2005;46:627-631
   percutaneous endoscopic jejunostomy
   prevent gastroesophageal reflux in patients      Martino R, Pron G, Diamant N. Screening for
   with preexisting esophagitis? Am J                 oropharyngeal dysphagia in stroke:
   Gastroenterol 2000; 9512):3439-3443.               insufficient evidence for guidelines.
                                                      Dysphagia 2000;15:19-30.
Lim SH, Lieu PK, Phua SY, Seshadri R,
   Venketasubramanian N, Lee SH et al.              Martino R, Silver F, Teasell R, et al. The
   Accuracy of bedside clinical methods               Toronto Bedside Swallowing Screening Test

15.   Dysphagia                                                                       pg. 57 of 60
  (TOR-BSST): development and validation of               neurological dysphagia. Br Med J
  a dysphagia screening tool for patients with            1992;304:1406-1409.
  stroke. Stroke 2009;40:555-561.
                                                       Park CL, O'Neill PA, Martin DF. A pilot
Meadows J. Dysphagia in unilateral cerebral              exploratory study of oral electrical
  lesions. J Neurol Neurosurg Psychiatry                 stimulation on swallow function following
  1973;36:853-860.                                       stroke: an innovative technique. Dysphagia
Milazzo LS, Bouchard J, Lund DA. The
   swallowing process: effects of aging and            Perez I, Smithard DG, Davies H, Kalra L.
   stroke. Physical Medicine and                         Pharmacological treatment of dysphagia in
   Rehabilitation: State of the Art Reviews              stroke. Dysphagia 1998;13:12-16.
                                                       Perlman PW, Cohen MA, Setzen M, et al. The
Muller-Lissner SA, Fimmel CJ, Will N, et al.             risk of aspiration of pureed food as
  Effect of gastric and transpyloric tubes on            determined by flexible endoscopic evaluation
  gastric emptying and duodenogastric reflux.            of swallowing with sensory testing.
  Gastroenterology 1982;83:1276-1279.                    Otolaryngol Head Neck Surg 2004;130:80-
Nakajoh K, Nakagawa T, Sekizawa K, Matsui T,
  Arai H, Sasaki H. Relation between incidence         Permsirivanich W, Tipchatyotin S, Wongchai M,
  of pneumonia and protective reflexes in                et al. Comparing the effects of rehabilitation
  post-stroke patients with oral or tube                 swallowing therapy vs. neuromuscular
  feeding. J Intern Med 2000;247:39-42.                  electrical stimulation therapy among stroke
                                                         patients with persistent pharyngeal
Nilsson H, Ekberg O, Olsson R, Hindfelt B.               dysphagia: a randomized controlled study. J
   Dysphagia in stroke: a prospective study of           Med Assoc Thai 2009;92:259-265.
   quantitative aspects of swallowing in
   dysphagic patients. Dysphagia 1998;                 Perry L. Screening swallowing function of
   13(1):32-38.                                          patients with acute stroke. Part one:
                                                         Identification, implementation and initial
Nishiwaki K, Tsuji T, Liu M, Hase K, Tanaka N,           evaluation of a screening tool for use by
   Fujiwara T. Identification of a simple                nurses. J Clin Nurs 2001; 10(4):463-473.
   screening tool for dysphagia in patients with
   stroke using factor analysis of multiple            Poels BJ, Brinkman-Zijlker HG, Dijkstra PU,
   dysphagia variables. J Rehabil Med 2005;              Postema K. Malnutrition, eating difficulties
   37(4):247-251.                                        and feeding dependence in a stroke
                                                         rehabilitation centre. Disabil Rehabil
Norton B, Horner-Ward M, Donnelly MT, Long               2006;28:637-643.
  RG, Holmes GK. A randomized prospective
  comparison of percutaneous endoscopic                Power ML, Fraser CH, Hobson A, et al.
  gastrostomy and nasogastric tube feeding               Evaluating oral stimulation as a treatment
  after acute dysphagic stroke. BMJ                      for dysphagia after stroke. Dysphagia
  1996;312: 13-16.                                       2006;21:49-55

Odderson IR, Keaton JC, McKenna BS.                    Ramsey DJ, Smithard DG, Kalra L. Early
  Swallow management in patients on an                   assessments of dysphagia and aspiration
  acute stroke pathway: quality is cost                  risk in acute stroke patients. Stroke 2003;
  effective. Arch Phys Med Rehabil                       34(5):1252-1257.
                                                       Ramsey DJ, Smithard DG, Kalra L. Can pulse
Paciaroni M, Mazzotta G, Corea F, Caso V,                oximetry or a bedside swallowing
  Venti M, Milia P et al. Dysphagia following            assessment be used to detect aspiration
  Stroke. Eur Neurol 2004; 51(3):162-167.                after stroke? Stroke 2006;37:2984-2988.

Park RHR, Allison MC, Lang J, Spence E, Morris         Robbins J, Levine RL, Maser A, Rosenbek JC,
  AJ, Danesh BJZ, Russel RJ, Mills PR.                   Kempster GB. Swallowing after unilateral
  Randomized comparison of percutaneous                  stroke of the cerebral hemisphere. Arch
  endoscopic gastrostomy and nasogastric                 Phys Med Rehabil 1993;74:1295-1300.
  tube feeding in patients with persisting

15.   Dysphagia                                                                           pg. 58 of 60
Robbins J, Kays SA, Gangnon RE et al. The             Sherman B, Nisenboum JM, Jesberger BL,
  effects of lingual exercise in stroke patients        Morrow CA, Jesberger JA. Assessment of
  with dysphagia. Arch Phys Med Rehabil.                dysphagia with the use of pulse oximetry.
  2007;88:150-158.                                      Dysphagia 1999; 14(3):152-156.

Rosenbek JC, Robbins J, Fishback B, Levine RL.        Shimizu T, Fujioka S, Otonashi H, Kondo M,
  Effects of thermal application on dysphagia           Sekizawa K. ACE inhibitor and swallowing
  after stroke. J Speech Hear Res                       difficulties in stroke. A preliminary study. J
  1991;34:1257-1268.                                    Neurol 2008;255:288-289.

Rosenbek JC, Robbins J, Willford WO, Kirk G,          Silver F, Norris J, Lewis A, Hachinski V. Early
  Schiltz A, Sowell TW, Deutsch SE, Milanti FJ,          mortality following stroke: a retrospective
  Ashford J, Gramigna GD, Fogarty A, Dong K,             review. Stroke 1984;15(3):492-496.
  Rau MT, Prescott TE, Lloyd AM, Sterkel MT,
  Hansen JE. Comparing treatment intensities          Smith HA, Lee SH, O'Neill PA, Connolly MJ. The
  of tactile-thermal application. Dysphagia             combination of bedside swallowing
  1998;13:1-9.                                          assessment and oxygen saturation
                                                        monitoring of swallowing in acute stroke: a
Rowat AM, Wardlaw JM, Dennis MS, Warlow                 safe and humane screening tool. Age Ageing
  CP. Does feeding alter arterial oxygen                2000; 29(6):495-499.
  saturation in patients with acute stroke?
  Stroke 2000; 31(9):2134-2140.                       Smithard DG, O'Neill PA, Park C, Morris J,
                                                        Wyatt R, England R, Martin DF.
Ryu JS, Park SR, Choi KH. Prediction of                 Complications and outcome after acute
  laryngeal aspiration using voice analysis. Am         stroke. Does dysphagia matter? Stroke
  J Phys Med Rehabil 2004;83:753-757                    1996;27:1200-1204.

Scmidt EV, Smirnov VE, Ryabova VS. Results            Smithard DG, Smeeton NC, Wolfe CD. Long-
  of the seven-year prospective study of                term outcome after stroke: does dysphagia
  stroke patients. Stroke 1988;19(8):942-               matter? Age Ageing. 2007;36:90-94.
                                                      Splaingard ML, Hutchins B, Sulton LD,
Schmidt J, Holas M, Halvorson K, Reding M.              Chaudhuri G. Aspiration in rehabilitation
  Videofluroscopic evidence of aspiration               patients: videofluoroscopy vs bedside
  predict pneumonia and death but not                   clinical assessment. Arch Phys Med Rehab
  dehydration following stroke. Dysphagia               1988;69:637-640.
                                                      Steele CM. Emergency room assessment and
Sellars C, Dunnet C, Carter R. A preliminary            intervention for dysphagia: a pilot project. J
  comparison of videofluoroscopy of swallow             Speech-Language Pathology and Audiology
  and pulse oximetry in the identification of           2002;26:100-110
  aspiration in dysphagic patients. Dysphagia
  1998; 13(2):82-86.                                  Teasell RW, Bach D, McRae M. Prevalence and
                                                        recovery of aspiration poststroke: a
Sellars C, Bowie L, Bagg J, et al. Risk factors         retrospective analysis. Dysphagia
  for chest infection in acute stroke: a                1994;9(1):35-39.
  prospective cohort study. Stroke
  2007;38:2284-2291                                   Teasell RW, Marchuk Y, McRae M, Finestone
                                                        HM. Pneumonia associated with aspiration
Seki T, Iwasaki K, Arai H, Sasaki H, Hayashi H,         following stroke. Arch Phys Med Rehabil
  Yamada S et al. Acupuncture for dysphagia             1996;77:707-709.
  in poststroke patients: a videofluoroscopic
  study. J Am Geriatr Soc 2005; 53(6):1083-           Teramoto S, Fukuchi Y. Detection of aspiration
  1084.                                                 and swallowing disorder in older stroke
                                                        patients: simple swallowing provocation test
Sharma JC, Fletcher S, Vassallo M, Ross I.What          versus water swallowing test. Arch Phys Med
  influences outcome of stroke--pyrexia or              Rehabil 2000;81:1517-1519.
  dysphagia? Int J Clin Pract 2001 ;55(1):17-
  20.                                                 Terre R, Mearin F. Oropharyngeal dysphagia
                                                        after the acute phase of stroke: predictors

15.   Dysphagia                                                                           pg. 59 of 60
  of aspiration. Neurogastroenterol Motil            Wang TG, Chang YC, Chen SY, Hsiao TY. Pulse
  2006; 18(3):200-205.                                 oximetry does not reliably detect aspiration
                                                       on videofluoroscopic swallowing study. Arch
Tobin MJ. Aspiration pneumonia. In: Dantzker           Phys Med Rehabil. 2005 Apr;86(4):730-4.
  DR (ed). Cardiopulmonary Critical Care.
  New York, Grune and Stratton, 1986.                Wanklyn P, Cox N, Belfield P. Outcome in
                                                       patients who require a gastrectomy after
Trapl M, Enderle P, Nowotny M, et al.                  stroke. Age and Aging 1995;24:510-514.
  Dysphagia bedside screening for acute-
  stroke patients: the Gugging Swallowing            Warnecke T, Teismann I, Meimann W, et al.
  Screen. Stroke 2007;38:2948-2952                     Assessment of aspiration risk in acute
                                                       ischaemic stroke--evaluation of the simple
Turner WW Jr. Nutritional considerations in            swallowing provocation test. J Neurol
  the patient with disabling brain disease.            Neurosurg Psychiatry 2008;79:312-314.
  Neurosurgery 1985;707-713.
                                                     Warnecke T, Teismann I, Oelenberg S, et al.
Turner-Lawrence DE, Peebles M, Price MF,               The safety of fiberoptic endoscopic
  Singh SJ, Asimos AW. A feasibility study of          evaluation of swallowing in acute stroke
  the sensitivity of emergency physician               patients. Stroke 2009;40:482-486.
  Dysphagia screening in acute stroke
  patients. Ann Emerg Med 2009;54:344-8,             Westergren A. Detection of eating difficulties
  348.                                                 after stroke: a systematic review. Int Nurs
                                                       Rev 2006; 53(2):143-149.
Veis S, Logemann J. Swallowing disorders in
  persons with cerebrovascular accidents.            Wu MC, Chang YC, Wang TG, Lin LC.
  Arch Phys Med Rehabil 1985;66:373-374.               Evaluating Swallowing Dysfunction Using a
                                                       100-ml Water Swallowing Test. Dysphagia
Wade DT, Hewer RL. Motor loss and swallowing           2004;19:43-47.
  difficulty after stroke: frequency, recovery,
  and prognosis. Acta Neurol Scand 1987;

15.   Dysphagia                                                                         pg. 60 of 60