CHEST Special Features

The Incidence of Dysphagia Following

Endotracheal Intubation
A Systematic Review
Stacey A. Skoretz, MSc; Heather L. Flowers, MEd, MHSc;
and Rosemary Martino, MA, PhD

Hospitalized patients are often at increased risk for oropharyngeal dysphagia following pro-
longed endotracheal intubation. Although reported incidence can be high, it varies widely. We
conducted a systematic review to determine: (1) the incidence of dysphagia following endotra-
cheal intubation, (2) the association between dysphagia and intubation time, and (3) patient char-
acteristics associated with dysphagia. Fourteen electronic databases were searched, using
keywords dysphagia, deglutition disorders, and intubation, along with manual searching of jour-
nals and grey literature. Two reviewers, blinded to each other, selected and reviewed articles at
all stages according to our inclusion criteria: adult participants who underwent intubation and
clinical assessment for dysphagia. Exclusion criteria were case series (n , 10), dysphagia deter-
mined by patient report, patients with tracheostomies, esophageal dysphagia, and/or diagnoses
known to cause dysphagia. Critical appraisal used the Cochrane risk of bias assessment and
Grading of Recommendations, Assessment, Development and Evaluation tools. A total of 1,489
citations were identified, of which 288 articles were reviewed and 14 met inclusion criteria. The
studies were heterogeneous in design, swallowing assessment, and study outcome; therefore, we
present findings descriptively. Dysphagia frequency ranged from 3% to 62% and intubation dura-
tion from 124.8 to 346.6 mean hours. The highest dysphagia frequencies (62%, 56%, and 51%)
occurred following prolonged intubation and included patients across all diagnostic subtypes. All
studies were limited by design and risk of bias. Overall quality of the evidence was very low. This
review highlights the poor available evidence for dysphagia following intubation and hence the
need for high-quality prospective trials. CHEST 2010; 137(3):665–673

Abbreviations: CSE 5 clinical swallowing evaluation; FEES 5 fiberoptic endoscopic evaluation of the swallow;
GRADE 5 Grading of Recommendations, Assessment, Development and Evaluation; VFS 5 videofluoroscopic swallow-
ing study

Endotracheal intubation and ventilatory support

are life-sustaining procedures often required dur-
the swallow physiology of the upper aerodigestive
tract. It occurs frequently in patients with structural or
ing the course of a patient’s hospitalization, but their neurologic disruption to the head and neck area from
presence can complicate resumption of oral intake.1,2 diseases such as stroke,5 head and neck cancer,6 and/or
Artificial airways often have negative effects on laryn- necessary medical treatments, including cervical spine
geal competence and overall swallowing physiology.1,2 surgery,7 prolonged intubation,3,4 tracheotomy,2,8 and
It remains to be determined whether oropharyngeal
dysphagia, if present, is attributed to the artificial For editorial comment see page 509
airway alone or in part to the underlying medical con-
ditions that precipitated its placement. Notwithstand- mechanical ventilation.1,2,9 Although dysphagia is itself
ing this uncertainty, patients on prolonged ventilation not a disease but rather a symptom of another medical
compose a diagnostic group at increased risk for condition and/or the interventions required to treat
oropharyngeal dysphagia.1-4 the condition, it can lead to a variety of medical com-
Oropharyngeal dysphagia, also referred to as dys- plications. Common consequences of dysphagia
phagia or disordered swallowing, is an abnormality of include dehydration; malnutrition10,11; aspiration of

www.chestpubs.org CHEST / 137 / 3 / MARCH, 2010 665

Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on March 5, 2010

© 2010 American College of Chest Physicians
oral secretions,12 food, or fluid13,14; and eventually Operational Definitions
death.10 Aspiration often leads to pneumonia.5,14-17 In We operationalized relevant terms a priori. Endotracheal intu-
fact, the risk of developing pneumonia is 11 times bation was defined as the presence of an endotracheal tube in the
greater in patients who aspirate compared with simi- oropharynx. Oropharyngeal dysphagia was defined as any impair-
lar patients with no aspiration.5 ment or abnormality of the oral, pharyngeal, or upper esophageal
Although the literature reports a high incidence of stage of deglutition. The presence of oropharyngeal dysphagia
was identified by either a bedside clinical swallow evaluation
dysphagia following intubation, these reports vary (CSE) or instrumental assessment, including videofluoroscopic
widely from 3%18 to 83%.1 Studies have also shown that swallow study (VFS) or fiberoptic endoscopic evaluation of the
prolonged intubation can be an independent predictor swallow (FEES).
of dysphagia.19,20 Artificial airways increase the risk
of upper airway injury and concomitant laryngeal Search Strategy
pathologies,21-23 which in turn affect upper airway From the start of online availability to May 2009, we searched
mechanics, aerodynamics, and protective reflexes.1,2 for eligible citations in 14 electronic databases (MEDLINE
These laryngeal pathologies include, but are not lim- [1950-], EMBASE [1980-], CINAHL [1982-], PsycINFO [1960-],
ited to, epithelial/mucosal abrasions, tracheoesopha- AMED [1985-], HealthSTAR [1966-], BIOSIS Previews [1980-],
geal fistula formation, tracheal stenosis, and granulation Cochrane DSR [1988-], ACP Journal Club [1991-], DARE
[1991-], CCTR [1991-], CMR [1995-], HTA [2001-], and
tissue.22,23 Multiple ventilation cycles can further exacer- NHSEED [1995-]) using the search terms deglutition disorders,
bate the dysphagia by disrupting the delicate synchrony swallowing disorders, dysphagia, swallowing, and intubation.
between swallowing and breathing, leading to aspira- In addition, we manually searched for relevant citations in 20
tion.9 Although the cause of dysphagia in these patients content-related journals between 1988 and May 2009, conference
is multifactorial and perhaps debatable,24,25 it is clear proceedings, and gray/unpublished literature (GrayLIT Network,
GraySource, OpenSigle, and Proquest Dissertations). We also
from the available literature that artificial airways reviewed citations from the accepted articles. A complete list of
interfere with the ability to execute an efficient and manually searched journals and conference proceedings is avail-
safe swallow. What is not yet known is how frequently able from the authors upon request.
dysphagia occurs and how to avoid its ill effects.
In order to evaluate the available evidence and Eligibility Criteria
attempt to resolve these uncertainties, we conducted
Of the identified citations, we included only articles with
a systematic review to assess (1) the incidence of dys- abstracts and those reporting the presence or absence of oropharyn-
phagia following intubation across various patient geal dysphagia in adult patients (. 18 years old) who underwent
diagnostic groups, (2) the association between dys- endotracheal intubation during their hospitalization. We accepted
phagia and the duration of endotracheal intubation, and retrospective or prospective study designs using only consecutive
(3) patient characteristics associated with dysphagia. enrollment, provided that the sample size exceeded 10. We
included articles published in any language. Specific to this study,
we defined swallowing assessment method to be clinical or instru-
Materials and Methods mental assessment. In order to avoid overestimating dysphagia
incidence secondary to endotracheal intubation, we excluded
A detailed protocol directed the various stages of our search articles with patients at high risk for dysphagia secondary to their
and appraisal of the literature on dysphagia and endotracheal primary diagnosis. These included patients with neurogenic or
intubation. head and neck diagnoses as well as tracheostomized patients.
Articles using only patient report to identify dysphagia were also
Manuscript received July 31, 2009; revision accepted October 13, excluded.
2009.
Affiliations: From the Department of Speech-language Pathology Study Selection
(Ms Skoretz and Drs Martino and Flowers), University of Toronto;
and the Toronto Western Research Institute, University Health The first two authors, blinded to each other’s results,
Network (Dr Martino), Toronto, Canada. reviewed all citations, abstracts, and articles to determine eligi-
Parts of this work were presented at the 2009 Dysphagia Research bility for inclusion. If the reviewers could not determine
Society Annual Meeting, New Orleans, LA, and the 2008 annual inclusion/exclusion based on the abstract alone, the citation was
American Speech Language and Hearing Association Convention,
Chicago, IL. accepted. Articles of all accepted citations were retrieved and
Funding/Support: Ms Skoretz and Dr Flowers were supported reviewed to determine the final studies for inclusion. Disagree-
by Ontario Graduate Scholarships and the University of Toronto ments at all stages of the selection process were resolved by
Open Fellowships. Dr Martino was supported by a Canadian consensus.
Institutes of Health Research New Investigator Fellowship Award
in Aging. Assessment of Methodological Quality
Correspondence to: Stacey Skoretz, MSc, University of Toronto,
160-500 University Ave, Toronto, Ontario, M5G 1V7 Canada; The same two reviewers, once again blinded to each other’s
e-mail: stacey.skoretz@utoronto.ca
© 2010 American College of Chest Physicians. Reproduction results, assessed the included studies for risk of bias and quality
of this article is prohibited without written permission from the using the risk of bias assessment tool and the Grading of Recom-
American College of Chest Physicians (www.chestpubs.org/ mendations, Assessment, Development, and Evaluation (GRADE)
site/misc/reprints.xhtml). approach as recommended by the Cochrane Collaboration.26 All
DOI: 10.1378/chest.09-1823 disagreements were resolved by consensus.

666 Special Features

Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on March 5, 2010

© 2010 American College of Chest Physicians
 The risk of bias assessment tool26 included sequence genera- dysphagia. A total of 50 citations (45 articles, and five
tion, allocation concealment, blinding, consecutive enrollment, abstracts) were retrieved from manual searches. Six
and the completeness of outcome reporting. Study quality using
the GRADE approach gave the highest rating (high-level evidence)
studies were accepted following manual searching:
to randomized trials, whereas observational studies were rated three articles18,20,36 from content-related journals, two
as low in quality as suggested by the Cochrane Collaboration.26 articles37,38 from the bibliographic references of
Anticipating a large number of observational studies, we modified accepted studies, and one article39 from review of gray
the Cochrane quality rating by downgrading in the presence of literature databases. In the end, a total of 14 articles were
certain factors, including limitations in study design, indirectness
of evidence (eg, indirect population or study’s main outcomes),
accepted and underwent further analysis (Table 1).
unexplained heterogeneity, or the imprecision of results. Con- A comprehensive list of articles not accepted for full
versely, study quality was upgraded if study design and results review is available from the authors on request.
suggested little to no evidence of bias.
Study Characteristics and Quality Assessment
Data Extraction
Of the 14 accepted articles (Table 1), 11 were pro-
One reviewer using a form determined a priori extracted data spective, including two randomized trials,40,41 three
regarding study design, sample size, patient diagnoses, incidence
of dysphagia, method of dysphagia assessment, intubation dura-
cohort studies,4,37,38 one case-control study,32 and five
tion, and patient comorbidities. A second reviewer checked all case series.3,18,19,36,42 Of the retrospective studies, two
extracted data for accuracy. Because of the heterogeneity of articles20,43 were case series and one article39 was a
patient diagnoses, study methodology, and outcomes across cohort design. Patient diagnoses varied across studies.
accepted studies, we summarized the results descriptively. All A total of eight studies18-20,36-38,40,43 enrolled surgical
risk of bias analyses were conducted using the Cochrane Collabo-
ration software program Review Manager (RevMan, version
patients, and of these, five articles18-20,38,43 enrolled
5.0.20; The Nordic Cochrane Centre; Copenhagen, Denmark). cardiovascular surgery patients and the remaining
Quality assessment ratings were adapted from GRADEprofiler three articles36,37,40 enrolled patients with mixed surgi-
(GRADEpro, version 3.2.2; available from http://www. cal diagnoses. Three other studies4,32,39 enrolled patients
gradeworkinggroup.org). with mixed medical diagnoses. An additional three
studies3,41,42 enrolled patients with a variety of both
medical and surgical diagnoses.
Results According to Cochrane methodology,26 we assessed
Literature Retrieved each study for risk of bias and poor study design to
generate a quality GRADE (Table 2). Of the 14
We retrieved 1,489 citations through database and
included studies, one study42 had the lowest risk of bias,
manual searching (Fig 1). Of these, 351 did not have
with only one area receiving a rating of unclear. Of
abstracts and were eliminated. We reviewed the
the three randomized trials, one study41 had adequate
remaining 1,138 titles and abstracts. An additional 848
sequence generation and allocation concealment.
abstracts were eliminated because they were case
Only one other study40 declared blinding. All 14 stud-
series with sample sizes of , 10, did not enroll patients
ies accounted for their outcomes, but only eight spe-
consecutively, included pediatric patients, did not
cifically declared consecutive enrollment.3,4,18-20,36,42,43
report swallowing outcomes, or included patients
Outcomes were operationally defined in four
following tracheotomy. Two articles were not
studies,4,20,41,42 whereas only eight studies3,4,32,37-40,42
retrievable.27,28 We retrieved and reviewed the full
conducted the same swallow assessment for all
text articles of the remaining 288 citations. Of these,
study enrollees. All studies received a high or unclear
14 languages were represented, including English.
risk of bias rating in at least one area. Additionally,
Following full article review, 274 articles were
each study had factors that decreased the quality
eliminated because they did not meet our inclusion
of the evidence such as insensitive swallowing
criteria. Of those, 58 articles used only patient report
assessment measures,32,37,38,40 small sample size of
of dysphagic symptoms, 31 articles included patients
fewer than 50 enrollees,3,32,37,39,40,42 and different types
with esophageal diagnoses, 27 articles included
of swallowing assessment for enrollees.18-20,36,43 As a
patients with primary diagnoses of head and neck
result, each study in this review was assigned a
cancer and/or neurogenic diagnoses (eg, stroke, trau-
GRADE of very low.
matic brain injury, and neurosurgical patients), and
eight articles did not describe their method for Dysphagia Following Endotracheal Intubation
assessing the swallow. Other article eliminations
included three duplicate publications29-31 using Duration of Intubation and the Frequency Dysphagia:
patients from studies accepted for this review,3,32 two Seven of the included studies3,4,19,20,41-43 reported mean
articles33,34 with study samples based on only durations of intubation in those with and without dys-
patients referred for suspected dysphagia, and one phagia (Table 3). Leder and colleagues42 reported
article35 that enrolled only patients with confirmed the lengthiest intubation duration in the dysphagic

www.chestpubs.org CHEST /137 / 3 / MARCH, 2010 667

Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on March 5, 2010

© 2010 American College of Chest Physicians
 Figure 1. Study selection process.

patients (mean, 346.6 6 298.6 h) with a dysphagia enrollees with prolonged intubation defined as
frequency of 45%. Ajemian and colleagues3 reported greater than 48 h.
the highest dysphagia frequency with a mean intuba-
tion duration of 192.0 h in patients with dysphagia. Swallowing Assessment Methods: The methods
One study41 reported longer intubation durations used to assess swallowing function were variable.
in patients without dysphagia (mean, 288.0 6 235.2 h) All studies but one39 used instrumental methods
compared with those with dysphagia (mean, 254.4 6 to determine the presence or absence of dysphagia
175.2 h). Five of these studies3,4,20,41,42 included only and/or aspiration. Seven studies3,4,32,37,38,40,42 conducted

668 Special Features

Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on March 5, 2010

© 2010 American College of Chest Physicians
 Table 1—Study Characteristics and Frequency of Dysphagia According to Patient Diagnosis

Timing of Assessment Dysphagia

Study Study Design No.a Swallowing Assessment Method Postextubation Frequency
Surgical, cardiovascular
only
Barker et al20 Case series 254 Screening, CSE, and/or VFS NR 130/254 (51%)
Burgess et al38 Cohort 64 Chest radiographb Immediately to 8 h 13/64 (20%)c
Ferraris et al18 Case series 1,042 Screening, VFS NR 31/1,042 (3%)d
Hogue et al19 Case series 844 Screening, barium cineradiography NR 28/844 (3%)d
Rousou et al43 Case series 838 Screening, barium cineradiography NR 23/838 (3%)d
Surgical mixede
Davis and Cullen37 Cohort 26 Chest radiographb 10 and 15 min 9/26 (35%)c
Keeling et al36 Case series 133 CSE, VFS Within 48 h 19/133 (14%)c
Stanley et al40 Randomized 40 Chest radiographb NR 1/40 (3%)c
Mixed medicalf
de Larminat et al32 Case control 34 Swallow latency measurements Immediately 21/34 (62%)g
El Solh et al4 Cohort 84 FEES Within 48 h 37/84 (44%)c
Padovani et al39 Cohort 23 CSE 1-5 d 8/23 (35%)
Mixed medical-surgicalh
Ajemian et al3 Case series 48 FEES Within 48 h 27/48 (56%)c
Barquist et al41 Randomized 70 CSE and FEES 48 6 2 hi, 24 6 2 hj 7/70 (10%)c
Leder et al42 Case series 20 FEES 24 6 2 h 9/20 (45%)c
CSE 5 clinical swallowing evaluation; FEES 5 fiberoptic endoscopic evaluation of the swallow; NR 5 not reported; VFS 5 videofluoroscopic
swallowing study.
aIncludes only patients meeting inclusion criteria for this review.

bWith administration of oral contrast agent.

cDysphagia defined as aspiration only.

dInstrumental assessment conducted with only those patients failing dysphagia screening.

eIncludes limb arthroplasty, thoracotomy or pulmonary resection, abdominal or vascular surgery.

fIncludes patients with respiratory illnesses, sepsis, liver failure, and/or other medical illness.

gDysphagia defined as swallowing latency on day 0.

hIncludes surgical/medical intensive care patients, critically ill trauma, burns, and/or elective surgical patients.

iCSE only.

jFEES only.

instrumentation on all study enrollees. Three studies3,4,42 VFS or barium cineradiography only on those
used FEES, three studies37,38,40 used chest radiogra- patients who failed swallowing screening. CSE was the
phy following administration of an oral contrast agent, sole method used to assess dysphagia in one study,39
and one study32 measured swallowing latency via sub- whereas another study41 used either CSE or FEES,
mental electromyography. Three studies18,19,43 used depending on the arm of their randomized trial.

Table 2—Risk of Bias and Methodologic Quality Across Studies

Outcomes Consistent
Sequence Allocation Consecutive All Outcomes Operationally Assessment for All GRADE
Study Generation Concealment Blinding Enrollment Addressed Defined Enrollees Strength
Ajemian et al3 N/A N/A Unclear Yes Yes Unclear Yes Very low
Barker et al20 N/A N/A Unclear Yes Yes Yes No Very low
Barquist et al41 Yes Yes No N/A Yes Yes N/A Very low
Burgess et al38 Unclear Unclear Unclear Unclear Yes No Yes Very low
Davis and Cullen37 N/A N/A Unclear Unclear Yes No Yes Very low
deLarminat et al32 N/A N/A Unclear Unclear Yes No Yes Very low
El Solh et al4 N/A N/A No Yes Yes Yes Yes Very low
Ferraris et al18 N/A N/A Unclear Yes Yes No Unclear Very low
Hogue et al19 N/A N/A No Yes Yes No Unclear Very low
Keeling et al36 N/A N/A Unclear Yes Yes No No Very low
Leder et al42 N/A N/A Unclear Yes Yes Yes Yes Very low
Padovani et al39 N/A N/A Unclear Unclear Yes No Yes Very low
Rousou et al43 N/A N/A Unclear Yes Yes No Unclear Very low
Stanley et al40 Unclear Unclear Yes Unclear Yes No Yes Very low
GRADE 5 Grading of Recommendations, Assessment, Development and Evaluation; N/A 5 not applicable.

www.chestpubs.org CHEST /137 / 3 / MARCH, 2010 669

Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on March 5, 2010

© 2010 American College of Chest Physicians
 Table 3—Intubation Duration According to Presence of Dysphagia

Intubation Duration, h
(mean 6 SD)a

Study Targeted Intubation Time, h Dysphagia No Dysphagia Dysphagia Frequency, %

Leder et al42 . 48 346.6 6 298.6 283.7 6 192.0 45
Barquist et al41 . 48 254.4 6 175.2 288.0 6 235.2 10
Rousou et al43 NR 200.9 6 75.0 15.3 6 1.6 3
El Solh et al4,b . 48 223.2 6 156.0 184.8 6 112.8 36
Ajemian et al3 . 48 192.0 184.8 56
El Solh et al4,c . 48 187.2 6 165.6 148.8 6 127.2 52
Barker et al20 . 48 142.4 6 63.0 87.1 6 43.3 51
Hogue et al19 NR 124.8 6 40.8 50.4 6 4.8 3
See Table 1 for expansion of abbreviation.
aWhere reported.

bYoung age cohort ( , 65 y old).

cElderly age cohort ( . 65 y old).

Across studies, swallowing assessments were con- some studies reported association with several risk
ducted at various time periods following extubation. factors, whereas others reported no association for
Six studies3,4,36,41,42 conducted their swallowing assess- the same risk factors (Table 5).
ment between 24 and 48 h following extubation.
Studies using chest radiographs37,38,40 administered
oral contrast at various time points with radiographs Discussion
taken at 2 min,40 30 min,38 and 1 h37 following the con- This systematic review verifies that reported dys-
trast ingestion. The study using electromyography32 phagia frequency following endotracheal intubation
measured swallow latency immediately (day 0), and is variable, ranging from 3%18,19,40,43 to 62%.32 More than
at 1, 2, and 7 days following extubation. One study39 one-half of the studies3,4,20,32,37-39,42 reported a dysphagia
assessed swallowing between days 1 and 5 following frequency exceeding 20%. The highest dysphagia
extubation. For another five studies18-20,40,43 the timing
of swallowing assessment was not reported.
Table 4—Surgical and Medical Risk Factor Association
Frequency of Dysphagia Following Intubation: With Dysphagia
The incidence of dysphagia across studies included in Associated with dysphagia
this review ranged widely from 3%18,19,40,43 to 62%.32 Congestive heart failure18,20
Those studies reporting the highest dysphagia Functional status4
frequencies,3,4,20,32,42 between 44% and 62%, had Increased hospital LOS18-20,43
prolonged intubation periods. The three studies18,19,43 Hypercholesterolemia18
Increased ICU LOS4,19
reporting the lowest dysphagia frequency did not Multiple intubations20
report findings from screening and/or CSE and only Increased operative time43
reported dysphagia in those patients with abnormal Perioperative TEE19,43
VFS or barium cineradiography. One study20 used Sepsis20
either CSE or VFS to determine dysphagia frequency Not associated with dysphagia
APACHE scores4,32
but did not stratify according to assessment method. COPD3,18-20
Swallowing impairment as an outcome was defined Circulatory shock18,20
differently across studies, either as any level of dys- Elevated CPB time18,19,43
phagia severity,18-20,39,43 only aspiration,3,4,36-38,40-42 or GERD3
only swallowing latency.32 Regardless of the defini- Hypertension18-20
ICU readmission20
tion, the dysphagia frequencies varied widely. Studies Myocardial infarction19,20
reporting on any level of dysphagia severity18-20,39,43 NYHA . 218,20
had frequencies ranging from 3%18,19,43 to 51%.20 Peripheral vascular disease18
Those studies reporting only aspiration also had wide- Preoperative CVA18-20
ranging frequencies from 3%40 to 56%.3 Smoking20
Surgery urgency20
Several of the included studies identified patient
risk factors, surgical and/or medical, associated with APACHE 5 Acute Physiology and Chronic Health Evaluation; CPB 5
cardiopulmonary bypass; CVA 5 cerebral vascular accident; GERD 5
dysphagia (Tables 4 and 5). Some risks were consis- gastroesophageal reflux disease; LOS 5 length of stay; NYHA 5 New
tently associated with dysphagia, whereas others were York Heart Association staging (heart failure); TEE 5 transesophageal
consistently not associated (Table 4). In contrast, echocardiography.

670 Special Features

Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on March 5, 2010

© 2010 American College of Chest Physicians
 Table 5—Studies Either Supporting or Not Supporting seriously question the accuracy of these frequency
Risks for Dysphagia estimates. In contrast, the three studies reporting the
Association with Dysphagia lowest dysphagia frequencies had the largest sample
sizes.18,19,43 We also question the accuracy of these
Risk Factors Supporting Not Supporting
frequency estimates because they appeared to include
Age Barquist et al 41 Ajemian et al3 patients with relatively short intubation durations who
Ferraris et al18 de Larminat et al32 consequently were not expected to have dysphagia.
Hogue et al19 Leder et al42
Rousou et al43
Across studies, the timing of swallowing assessments
Decreased cardiac output Hogue et al19 Ajemian et al 3 ranged from immediately32,38 to five days39 following
Diabetes mellitus Ferraris et al18 Barker et al20 extubation. In general, one would expect the highest
Ajemian et al3 dysphagia frequency in patients with prolonged intuba-
Barker et al20 tion durations and assessments conducted immediately
Hogue et al19
Intubation duration Barker et al20 Ajemian et al3
following extubation. However, due to the poor study
Hogue et al19 Barquist et al41 quality and variability of the included studies, our
Rousou et al43 de Larminat et al32 findings could not corroborate this premise.
El Solh et al4 A wide assortment of swallowing assessment meth-
Leder et al42 ods, including screening, clinical swallowing evalua-
Left ventricle ejection Rousou et al43 Hogue et al19
fraction
tions, and a variety of instrumental assessments, were
Perioperative CVA Rousou et al43 Barker et al20 included in the accepted studies. Dysphagia
Hogue et al19 frequency varied regardless of assessment type.
Postoperative pulmonary Hogue et al19 Leder et al42 Although heterogeneity across studies did not permit
complications
Preoperative/ Hogue et al19,a Barker et al20
a statistical association of dysphagia frequency and
postoperative IABP swallow assessment, studies using FEES on all
Renal risks Ferraris et al18 Barker et al20 enrollees reported some of the highest frequencies
Hogue et al19 of dysphagia from 44%4 to 56%.3 In contrast, other
Surgery type Barker et al20,b Hogue et al19
Ferraris et al18,c Rousou et al43
studies assessing the swallow with static chest radio-
Tube feeding Barker et al20 Ajemian et al3 graphs37,38,40 or clinical measures39 detected a lower
El Solh et al4 incidence of dysphagia, from 3%40 to 35%.37,39 When
Leder et al42 compared with other swallowing assessment meth-
IABP 5 intraaortic balloon pump. See Table 4 for expansion of other ods, direct visualization of pharyngeal and laryngeal
abbreviations.
aPostoperative IABP only.
swallowing structures (eg, FEES) may be a more sen-
bCoronary artery bypass.
sitive measure.44,45
cNoncoronary procedures. Studies varied in how swallowing outcomes
were defined. More than one-half of the included
studies3,4,36-38,40-42 used aspiration as their main swal-
frequencies of 62%,32 56%,3 and 51%20 included lowing outcome. Aspiration is only one aspect across
patients experiencing prolonged intubation ( . 24 h) the spectrum of swallowing impairments. Although
across all diagnostic subtypes, mixed medical, mixed aspiration is considered dysphagia in its most severe
medical-surgical, and cardiovascular surgical groups, form, defining dysphagia as such would limit diagnos-
respectively. Hence, no single diagnosis appeared to tic scope and potentially miss other significant swal-
be associated with greater risk of dysphagia. The wide lowing findings. For example, one study41 reporting a
range of dysphagia frequency identified in this review low incidence of dysphagia (10%) also commented
is more likely attributed to variations in study design, on other pharyngeal aspects of the swallow. These
such as method and timing of swallowing assessment. findings were reported only qualitatively, thereby
Many studies were observational and few declared blind- explaining, in part, their low reported frequency.
ing or operational definitions. Together, design variabil- Although we used stringent selection criteria and
ity and poor quality resulted in studies with a high risk rigorous methodology while excluding confounding
of bias, thereby weakening the available evidence on diagnoses for dysphagia, this review is limited by
dysphagia frequency following endotracheal intubation. many factors. Most limitations were imposed by the
Across all studies, poor study quality and high risk design, quantity, and quality of the original research.
of bias likely led to either underreporting or overre- The few included studies were heterogeneous, dif-
porting of dysphagia. Oddly, studies reporting the fering in regard to their outcomes, study design, and
longest intubation durations did not report the highest patient diagnoses. Consequently, instead of combin-
dysphagia frequencies.41-43 However, their large stan- ing outcome data with a metaanalysis, we chose to
dard deviations,41,42 coupled with failure to use the use descriptive methods. We found insufficient evi-
same instrumental assessments for all enrollees,43 dence to calculate the relative risk of dysphagia

www.chestpubs.org CHEST /137 / 3 / MARCH, 2010 671

Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on March 5, 2010

© 2010 American College of Chest Physicians
following a range of intubation durations or to 5. Martino R, Foley N, Bhogal S, Diamant N, Speechley M,
determine the effect of intubation duration on the Teasell R. Dysphagia after stroke: incidence, diagnosis, and
pulmonary complications. Stroke. 2005;36(12):2756-2763.
frequency of dysphagia across all studies. We propose 6. Ward EC, Bishop B, Frisby J, Stevens M. Swallowing out-
that using sensitive swallow assessments on all enroll- comes following laryngectomy and pharyngolaryngectomy.
ees, while reporting on all aspects of swallowing Arch Otolaryngol Head Neck Surg. 2002;128(2):181-186.
function, would best represent the frequency and 7. Smith-Hammond CA, New KC, Pietrobon R, Curtis DJ,
characteristics of dysphagia following extubation. Scharver CH, Turner DA. Prospective analysis of incidence
and risk factors of dysphagia in spine surgery patients: com-
Future research endeavors should include homo- parison of anterior cervical, posterior cervical, and lumbar
geneous patient populations or larger sample sizes procedures. Spine (Phila Pa 1976). 2004;29(13):1441-1446.
and rigorous methodology. This research is neces- 8. Bonanno PC. Swallowing dysfunction after tracheostomy.
sary to permit clinicians to identify patients who Ann Surg. 1971;174(1):29-33.
are at greater risk for dysphagia and enable more 9. Elpern EH, Scott MG, Petro L, Ries MH. Pulmonary aspira-
tion in mechanically ventilated patients with tracheostomies.
appropriate interventions. Chest. 1994;105(2):563-566.
In conclusion, there are relatively few studies with 10. Smithard DG, O’Neill PA, Parks C, Morris J. Complications
specific outcomes that focus on dysphagia following and outcome after acute stroke. Does dysphagia matter?
intubation. The majority of identified studies are Stroke. 1996;27(7):1200-1204.
of very low quality with high risk of bias. Although 11. Westergren A, Ohlsson O, Rahm Hallberg I. Eating difficul-
ties, complications and nursing interventions during a period
variable, most studies with prolonged intubation of three months after a stroke. J Adv Nurs. 2001;35(3):416-
durations and those that conducted instrumental 426.
assessments on the entire study population reported 12. Murray J, Langmore SE, Ginsberg S, Dostie A. The
higher frequencies of dysphagia. Given the high like- significance of accumulated oropharyngeal secretions and
lihood of serious medical complications of dysphagia, swallowing frequency in predicting aspiration. Dysphagia.
1996;11(2):99-103.
particularly pneumonia, we recommend that swal- 13. Smith CH, Logemann JA, Colangelo LA, Rademaker AW,
lowing assessments should be conducted on patients Pauloski BR. Incidence and patient characteristics associated
undergoing prolonged intubation durations. In the with silent aspiration in the acute care setting. Dysphagia.
meantime, we call for high-quality studies using 1999;14(1):1-7.
homogeneous patient populations to assess the influ- 14. Lundy DS, Smith C, Colangelo L, et al. Aspiration: cause and
implications. Otolaryngol Head Neck Surg. 1999;120(4):474-
ence of prolonged intubation on dysphagia and to 478.
determine whether select medical comorbidities put 15. Holas MA, DePippo KL, Reding MJ. Aspiration and relative
patients at greater risk. risk of medical complications following stroke. Arch Neurol.
1994;51(10):1051-1053.
16. Johnson ER, McKenzie SW, Sievers A. Aspiration pneumonia
in stroke. Arch Phys Med Rehabil. 1993;74(9):973-976.
Acknowledgments
17. Loeb M, McGeer A, McArthur M, Walter S, Simor AE. Risk
Financial/nonfinancial disclosures: The authors have reported factors for pneumonia and other lower respiratory tract
to CHEST that no potential conflicts of interest exist with any infections in elderly residents of long-term care facilities.
companies/organizations whose products or services may be Arch Intern Med. 1999;159(17):2058-2064.
discussed in this article.
18. Ferraris VA, Ferraris SP, Moritz DM, Welch S. Oropharyngeal
Other contributions: We thank those who donated their
multilingual abilities and time for the many translations required dysphagia after cardiac operations. Ann Thorac Surg.
for this study. We gratefully acknowledge the authors who 2001;71(6):1792-1795.
responded to our requests for further information. We also thank 19. Hogue CW Jr, Lappas GD, Creswell LL, et al. Swallowing
Drs Terrence Yau, John Granton, and Kevin O’Kelly for their dysfunction after cardiac operations. Associated adverse out-
valuable input throughout the course of this project. This work comes and risk factors including intraoperative transesoph-
originated at the University of Toronto. ageal echocardiography. J Thorac Cardiovasc Surg. 1995;
110(2):517-522.
20. Barker J, Martino R, Reichardt B, Hickey EJ, Ralph-Edwards
A. Incidence and impact of dysphagia in patients receiving
References
prolonged endotracheal intubation after cardiac surgery. Can
1. Tolep K, Getch CL, Criner GJ. Swallowing dysfunction in J Surg. 2009;52(2):119-124.
patients receiving prolonged mechanical ventilation. Chest. 21. Sellery GR, Worth A, Greenway RE. Late complica-
1996;109(1):167-172. tions of prolonged tracheal intubation. Can Anaesth Soc J.
2. DeVita MA, Spierer-Rundback L. Swallowing disorders in 1978;25(2):140-143.
patients with prolonged orotracheal intubation or tracheos- 22. Stauffer JL, Olson DE, Petty TL. Complications and conse-
tomy tubes. Crit Care Med. 1990;18(12):1328-1330. quences of endotracheal intubation and tracheotomy. A pro-
3. Ajemian MS, Nirmul GB, Anderson MT, Zirlen DM, Kwasnik spective study of 150 critically ill adult patients. Am J Med.
EM. Routine fiberoptic endoscopic evaluation of swallowing 1981;70(1):65-76.
following prolonged intubation: implications for manage- 23. Sue RD, Susanto I, Sue RD, et al. Long-term complications
ment. Arch Surg. 2001;136(4):434-437. of artificial airways. Clin Chest Med. 2003;24(3):457-471.
4. El Solh A, Okada M, Bhat A, Pietrantoni C. Swallowing disor- 24. Leder SB, Ross DA. Investigation of the causal relationship
ders post orotracheal intubation in the elderly. Intensive Care between tracheotomy and aspiration in the acute care setting.
Med. 2003;29(9):1451-1455. Laryngoscope. 2000;110(4):641-644.

672 Special Features

Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on March 5, 2010

© 2010 American College of Chest Physicians
25. Terk AR, Leder SB, Burrell MI. Hyoid bone and laryngeal for pulmonary resection. Ann Thorac Surg. 2007;83(1):
movement dependent upon presence of a tracheotomy tube. 193-196.
Dysphagia. 2007;22(2):89-93. 37. Davis FG, Cullen DJ. Post-extubation aspiration follow-
26. Higgins JPT, Green S, eds. Cochrane Handbook for Systematic ing prolonged intubation [abstract]. American Society of
Reviews of Interventions Version 5.0.1 [updated September Anesthesiologists Annual Meeting. Washington, DC: American
2008]. The Cochrane Collaboration. http://www.cochrane- Society of Anesthesiologists. 1974; 181-182.
handbook.org. Accessed July 1, 2008. 38. Burgess GE III, Cooper JR Jr, Marino RJ, Peuler MJ,
27. Vasilev D, Germanova E, Tzaner V, et al. Sources of complica- Warriner RA III. Laryngeal competence after tracheal extu-
tions in Laryngeal Mask Airway (LMA) insertion [Bulgarian]. bation. Anesthesiology. 1979;51(1):73-77.
Anaesth Intensive Care. 2005;32(5):19-22. 39. Padovani AR, Moraes DP, de Medeiros GC, et al. Orotracheal
28. Sarroca O, Mengual M, Pedro-Botet ML, et al. Descriptive intubation and dysphagia: comparison of patients with and
analysis of 124 cases of Legionnaire’s disease (LD) in without brain damage. einstein. 2008;6(3):343-349.
the elderly. Abstracts of the Interscience Conference on 40. Stanley GD, Bastianpillai BA, Mulcahy K, Langton JA.
Antimicrobial Agents and Chemotherapy. 2003;43:341. Postoperative laryngeal competence. The laryngeal mask air-
29. de Larminat V, Montravers P, Dureuil B, et al. Consequences way and tracheal tube compared. Anaesthesia. 1995;50(11):
of prolonged endotracheal intubation on swallowing reflex 985-986.
[abstract]. Eur J Anaesthesiol. 1991;8:487-488. 41. Barquist E, Brown M, Cohn S, Lundy D, Jackowski J.
30. de Larminat V, Dureuil B, Montravers P, Desmonts JM. Postextubation fiberoptic endoscopic evaluation of swallow-
Impairment of deglutition reflex after prolonged intubation ing after prolonged endotracheal intubation: a randomized,
[in French]. Ann Fr Anesth Reanim. 1992;11(1):17-21. prospective trial. Crit Care Med. 2001;29(9):1710-1713.
31. Laryea JA, Ajemian MS. Risk factors for prolonged swallow- 42. Leder SB, Cohn SM, Moller BA. Fiberoptic endoscopic
ing dysfunction following prolonged endotracheal intubation documentation of the high incidence of aspiration following
[abstract]. Chest. 2006;130(4):207S. extubation in critically ill trauma patients. Dysphagia. 1998;
32. de Larminat V, Montravers P, Dureuil B, Desmonts JM. 13(4):208-212.
Alteration in swallowing reflex after extubation in intensive 43. Rousou JA, Tighe DA, Garb JL, et al. Risk of dysphagia
care unit patients. Crit Care Med. 1995;23(3):486-490. after transesophageal echocardiography during cardiac
33. Leder SB, Suiter DM. Effect of nasogastric tubes on incidence operations. Ann Thorac Surg. 2000;69(2):486-489.
of aspiration. Arch Phys Med Rehabil. 2008;89(4):648-651. 44. Kelly AM, Drinnan MJ, Leslie P. Assessing penetration and
34. Leder SB, Sasaki CT, Burrell MI. Fiberoptic endoscopic eval- aspiration: how do videofluoroscopy and fiberoptic endo-
uation of dysphagia to identify silent aspiration. Dysphagia. scopic evaluation of swallowing compare? Laryngoscope.
1998;13(1):19-21. 2007;117(10):1723-1727.
35. Partik B, Pokieser P, Schima W, et al. Videofluoroscopy of 45. Kelly AM, Leslie P, Beale T, Payten C, Drinnan MJ.
swallowing in symptomatic patients who have undergone long- Fibreoptic endoscopic evaluation of swallowing and video-
term intubation. AJR Am J Roentgenol. 2000;174(5):1409-1412. fluoroscopy: does examination type influence perception of
36. Keeling WB, Lewis V, Blazick E, Maxey TS, Garrett JR, Sommers pharyngeal residue severity? Clin Otolaryngol. 2006;31(5):
KE. Routine evaluation for aspiration after thoracotomy 425-432.

www.chestpubs.org CHEST /137 / 3 / MARCH, 2010 673

Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on March 5, 2010

© 2010 American College of Chest Physicians