www.nature.

com/scientificreports

OPEN Predictors of post‑extubation

stridor in patients on mechanical
ventilation: a prospective
observational study
Aiko Tanaka1*, Akinori Uchiyama1, Yu Horiguchi1, Ryota Higeno2, Ryota Sakaguchi1,
Yukiko Koyama1, Hironori Ebishima1, Takeshi Yoshida1, Atsuhiro Matsumoto3, Kanaki Sakai1,
Daisuke Hiramatsu1, Naoya Iguchi1, Noriyuki Ohta4 & Yuji Fujino1

The cuff leak test (CLT) has been widely accepted as a simple and noninvasive method for predicting
post-extubation stridor (PES). However, its accuracy and clinical impact remain uncertain. We
aimed to evaluate the reliability of CLT and to assess the impact of pre-extubation variables on the
incidence of PES. A prospective observational study was performed on adult critically ill patients who
required mechanical ventilation for more than 24 h. Patients were extubated after the successful
spontaneous breathing trial, and CLT was conducted before extubation. Of the 191 patients studied,
26 (13.6%) were deemed positive through CLT. PES developed in 19 patients (9.9%) and resulted in
a higher reintubation rate (8.1% vs. 52.6%, p < 0.001) and longer intensive care unit stay (8 [4.5–14]
vs. 12 [8–30.5] days, p = 0.01) than patients without PES. The incidence of PES and post-extubation
outcomes were similar in patients with both positive and negative CLT results. Compared with
patients without PES, patients with PES had longer durations of endotracheal intubation and required
endotracheal suctioning more frequently during the 24-h period prior to extubation. After adjusting
for confounding factors, frequent endotracheal suctioning more than 15 times per day was associated
with an adjusted odds ratio of 2.97 (95% confidence interval, 1.01–8.77) for PES. In conclusion,
frequent endotracheal suctioning before extubation was a significant PES predictor in critically
ill patients. Further investigations of its impact on the incidence of PES and patient outcomes are
warranted.

Endotracheal intubation is an essential procedure for respiratory support in critically ill patients. However,
complications resulting from mechanical trauma to the larynx by the endotracheal tube may lead to laryngeal
­edema1,2. Airflow through a narrowed upper airway manifests clinically as a post-extubation stridor (PES)3. The
incidence of PES is 4–10% in patients who are endotracheally intubated for more than 24 ­h4,5. Patients with PES
are likely to undergo increased r­ eintubation6,7, which is associated with increased morbidity and m
­ ortality8–10.
To identify patients at risk of laryngeal edema, airway patency assessment before extubation has been evalu-
ated. Since Miller and Cole evaluated 100 mechanically ventilated patients in 1996, the cuff leak test (CLT) has
been widely used because of its noninvasiveness and avoidance of sophisticated ­equipment11–15. The cuff leak
volume was defined as the tidal volume difference from before to after endotracheal tube (ETT) cuff deflation. The
primary threshold of cuff leak volume of less than 110 mL was reported as the valid cut-off value for predicting
PES with a positive predictive value of 0.80. Subsequent studies reported various cut-off ­values16–18. The other
remarkable threshold, a cuff leak less than 10% of the tidal volume, was identified to predict PES or reintubation
with a specificity of 96%19. In the previous meta-analyses of pooled studies on multiple CLT methods, the CLT
displayed an accurate prediction for patients at high risk of the development of airway ­obstruction20–22 and is
now widely used. However, some previous studies did not conduct a spontaneous breathing trial (SBT), which

1
Department of Anesthesiology and Intensive Care, Osaka University Graduate School of Medicine, 2‑15
Yamadaoka, Suita, Osaka  565‑0871, Japan. 2Division of Pediatrics, Osaka General Medical Center, 3‑1‑56
Bandai‑Higashi, Sumiyoshi‑ku, Osaka 558‑8558, Japan. 3Division of Anesthesiology, Osaka General Medical Center,
3‑1‑56 Bandai‑Higashi, Sumiyoshi‑ku, Osaka 558‑8558, Japan. 4Department of Anesthesiology, Kindai University
Faculty of Medicine, 377‑2, Ohno‑Higashi, Osakasayama, Osaka  589‑8511, Japan. *email: aikotanakaicu@
gmail.com

Scientific Reports | (2021) 11:19993 | https://doi.org/10.1038/s41598-021-99501-8 1

Vol.:(0123456789)
 www.nature.com/scientificreports/

is the current standard weaning method, and there is concern that this may affect extubation outcomes. Indeed,
recent studies have reported the poor diagnostic performance of the CLT in the current critical care s­ ettings7,23,24,
and the diagnostic performance of CLT remains controversial.
Moreover, several studies related to baseline characteristics and airway management were performed to iden-
tify PES incidence risk factors. Important risk factors include female sex, body mass index (BMI), larger ETT
diameter, and longer intubation d ­ uration3,12,25–27. However, each risk factor is insufficient to identify patients at
high risk for PES when considered ­alone3, and the investigation of physiological or descriptive data at extuba-
tion is limited.
Accordingly, we aimed to determine the CLT value and the impact of pre-extubation variables for predict-
ing PES among critically ill patients, based on the current standard and unified weaning readiness techniques.

Methods
Conduct of the study and selection criteria.  This prospective study was performed in a multidisci-
plinary intensive care unit (ICU) of a tertiary care hospital between May 2017 and April 2019. The consecutive
patients who had undergone invasive mechanical ventilation for more than 24 h and had been extubated fol-
lowing a successful SBT and assessment of the CLT were enrolled. Patients younger than 18 years of age, with a
tracheostomy or unplanned extubation, who received palliative care or passed away under mechanical ventila-
tion, and who passed away within 48 h after extubation were excluded. If patients received mechanical ventila-
tion multiple times, the first extubation attempt after mechanical ventilation for more than 24 h was included.
This study was approved by the ethics review board of Osaka University Hospital (Approval Number: 19316),
and the need for consent for research participation was waived. The study was performed in accordance with
relevant guidelines and regulations.

Assessment before extubation.  According to the weaning strategies with national consensus, patients
were considered eligible for SBT when the underlying cause of respiratory failure was resolved. Sufficient oxy-
genation ­(SpO2 > 90% at ­FiO2 ≤ 0.5 and positive end-expiratory pressure [PEEP] ≤ 8 ­cmH2O) and inspiratory
effort (tidal volume > 5 mL/kg; minute volume < 15 L/min; rapid shallow breathing index < 105 breaths/min/L;
and pH > 7.25) were verified prior to SBT. Clinicians suspended SBT if a patient had any of the following: sig-
nificant deterioration of oxygenation compared to before SBT, respiratory rate ≥ 30 breaths/min, heart rate ≥ 140
beats/min, arrhythmia, myocardial ischemia symptom, sustained increased blood pressure, or appearance of
respiratory distress as defined by paradoxical breathing, use of accessory muscles, sweating, or agitation. Patients
were extubated if they tolerated 30 min of spontaneous breathing on PEEP of 5 c­ mH2O with pressure support
(PS) of 5 ­cmH2O.
The cuff pressure was continuously measured and maintained at 25 c­ mH2O during mechanical ventilation in
all patients. Prior to endotracheal extubation, CLT was performed by the attending intensivists. The patient was
suctioned intraorally and intratracheally and placed in the assist-control ventilation mode. With the ETT cuff
inflated to the occlusion volume, the average mechanical exhaled volume over the three consecutive respiratory
cycles was measured and recorded. Subsequently, the ETT cuff was deflated. The mechanical exhaled tidal volume
was monitored over the next six consecutive respiratory cycles to ensure the exclusion of erroneous values. The
average of the three lowest exhaled tidal volumes was recorded. The lost tidal volume on exhalation, the cuff leak
volume, was calculated as the tidal volume difference between the inflated and deflated ETT cuffs. Moreover,
the cuff leak volume was divided by the tidal volume before cuff deflation and multiplied by 1­ 0019; the resulting
value was regarded as the percent cuff leak. The high risk of upper airway obstruction was determined as the
positive cuff leak result, the cuff leak volume ≤ 110 mL, and/or the percent cuff leak ≤ 10%, before e­ xtubation11,19,21.

Data collection.  Variables collected of each patient included age, sex, BMI, comorbidities, Acute Physiol-
ogy and Chronic Health Evaluation (APACHE) II and III scores as indicators of disease severity on ICU admis-
sion, the reason for intubation, the size of the endotracheal tube, and the duration of mechanical ventilation. In
addition, arterial blood gas values and ventilation data during successful SBT were recorded. Simultaneously,
we collected information on patient management during the 24 h prior to extubation. Fluid balance, increased
body weight when compared with the body weight prior to ICU admission, and as an indicator of tracheal–bron-
chial secretion, endotracheal suctioning, were recorded 24 h prior to extubation. Systemic steroid therapy was
recorded if the patient was administered steroid ≥ 20 mg/day of methylprednisolone at least 4 h prior to extuba-
tion. Furthermore, the severity of organ dysfunction during extubation was assessed using the Sequential Organ
Failure Assessment (SOFA) ­score28.
Moreover, the following data were recorded after extubation: reintubation and the incidence of PES within
48 h after extubation, ICU and hospital lengths of stay, and ICU and hospital mortality. Patients’ respiratory func-
tion and hemodynamics were closely monitored for 48 h after extubation. They were reintubated if they presented
cardiac arrest, refractory hypoxemia, severe hemodynamic instability without response to fluids and vasoactive
drugs, persistent inability to remove excessive secretions, upper airway obstruction, agitation, and loss of con-
sciousness at the discretion of the intensivist in charge. A patient having PES was identified with the presence
of respiratory distress with high-pitched inspiratory wheezing. The incidence of PES was assessed and recorded
by clinicians and decisions concerning the medical treatment for PES were made by the attending intensivists.

Statistical analyses.  Categorical data are presented as numbers and percentages. Distributed data are pre-
sented as medians and interquartile ranges. Differences in proportions were evaluated using the chi-square test
or Fisher’s exact test, and differences in distributed data with the Mann–Whitney U test for the two groups. A
multiple logistic regression analysis model was used to determine PES incidence risk factors. Covariates with

Scientific Reports | (2021) 11:19993 | https://doi.org/10.1038/s41598-021-99501-8 2

Vol:.(1234567890)
www.nature.com/scientificreports/

Adult patients
who received mechanical ventilation
for more than 24 h
from May 2017 to April 2019
n = 286

Excluded:
Tracheostomy: n = 21
Palliative care or death under mechanical
ventilation: n = 9
Discharged with mechanical ventilation: n = 14
Unplanned extubation: n =1
Death within 48 h after the extubation: n = 0
Omission of the CLT: n = 50

Eligible study subjects

n = 191

Figure 1.  Flowchart of the study patients. CLT cuff leak test.

p < 0.05 in the univariate analysis were included in the multivariate model. To adjust for potential confounding
factors, sex, BMI (> 25 kg/m2), and positive CLT were added to the model. All statistical inferences were made
with a 2-sided significance level of 5% and conducted with R version 4.0.1 (2020, R Foundation for Statistical
Computing, Vienna, Austria).

Ethics declarations.  This study was approved by the ethics review board of Osaka University Hospital
(Approval Number: 19316), and the need for consent for research participation was waived.

Results
Over the 2-year study period, a total of 286 adult patients required invasive mechanical ventilation for more than
24 h. CLTs were not performed for 95 patients for the following reasons: tracheostomy (n = 21), palliative care
or death under mechanical ventilation (n = 9), ICU discharge with mechanical ventilation (n = 14), unplanned
extubation (n = 1), or omission of the CLT (n = 50). The remaining 191 adult patients were included in this study
(Fig. 1). The characteristics of the study population are presented in Table 1. The median age was 65 years, and 67
(35.1%) patients were women. The main causes of ICU admission were surgical (75.9%). Reasons for mechani-
cal ventilation included postoperative acute respiratory failure defined as ineligible decision for weaning and
extubation in patients after surgery (n = 143; 74.9%), pneumonia (n = 16; 8.4%), cardiac failure (n = 10; 5.2%),
and cardiac arrest (n = 8; 4.2%).

CLT result and its clinical impact.  Twenty-six (13.6%) patients were determined to be positive through
the CLT before extubation (Table 1). Positive CLT results were detected more often in female patients (18/67;
26.9%) than in male patients (8/124; 6.5%) (p < 0.001). Besides, patients with positive CLT results had a signifi-
cantly larger BMI than their counterparts. The ETT diameter, duration of mechanical ventilation, and data from
24 h before extubation were similar regardless of CLT results (Tables 1, 2). The cuff leak volume and percent cuff
leak were significantly smaller in patients with positive CLT results. All patients with positive CLT results had
a cuff leak volume ≤ 110 mL; however, only 9 (34.6%) patients had a percent cuff leak ≤ 10%. The PES incidence
was similar in the results of both CLT (negative: 9.7% vs. positive: 11.5%, p = 0.73) (Supplementary Table S1
shows this in more detail). Moreover, there were no differences in the reintubation rate, ICU and hospital lengths
of stay, and ICU and hospital mortality according to the CLT results.

Comparison of patients with and without PES.  The overall incidence of PES was 9.9% (19/191).
Table 3 compares the baseline characteristics of patients with and without PES. There were no significant differ-
ences in sex or BMI. The SOFA score at extubation, respiratory data, and the administration of systemic steroids
during the 24-h period before extubation were similar in both groups (Table 4). Patients with PES had a longer
duration of mechanical ventilation and required endotracheal suctioning more frequently than patients without
PES. However, the CLT results were similar in both groups. As post-extubation outcomes, 8 (42.1%) patients
with PES required reintubation for PES (Table 5). The reintubation rate for all causes was significantly higher in
patients with PES than in those without PES (8.1% vs. 52.6%, p < 0.001). Moreover, the ICU length of stay was
significantly longer for patients with PES than for those without PES (p = 0.01). There were no significant differ-
ences in hospital length of stay and ICU or hospital mortality.

Factors associated with PES.  In the univariate logistic regression analysis, the positive result of CLT was
not significantly associated with the incidence of PES, while there were differences in frequent endotracheal suc-
tioning and prolonged mechanical ventilation (Table 6). After adjusting for confounding factors in the multivari-
ate analysis, the frequent endotracheal suctioning more than 15 times during the 24-h period before extubation
remained significant (adjusted odds ratio 2.97; 95% confidence interval [CI] 1.01–8.77).

Scientific Reports | (2021) 11:19993 | https://doi.org/10.1038/s41598-021-99501-8 3

Vol.:(0123456789)
 www.nature.com/scientificreports/

All (n = 191) Negative CLT (n = 165) Positive CLT (n = 26) p value

Age, years 65 (53–74) 65 (53–74) 65 (50–74) 0.81
Sex (female) 67 (35.1%) 49 (29.7%) 18 (69.2%)  < 0.001
22.5 (20.0–25.5) 22.2 (19.9–25.2) 25.3 (22.5–26.9)
BMI, kg/m2 0.01
n = 189 n = 163 n = 26
61 (32.3%) 46 (28.2%) 15 (57.7%)
BMI > 25 kg/m2 0.01
n = 189 n = 163 n = 26
APACHE II score 15 (12–19) 15 (12–19) 15 (12–20) 0.55
APACHE III score 60 (48–71) 60 (48–71) 58.5 (51–73) 0.70
Comorbidities
   Heart failure 72 (37.7%) 66 (40%) 6 (23.1%) 0.13
   Asthma 7 (3.7%) 5 (3.0%) 2 (7.7%) 0.25
   COPD 8 (4.2%) 8 (4.8%) 0 (0%) 0.60
   Interstitial lung disease 12 (6.3%) 10 (6.1%) 2 (7.7%) 0.67
Surgical diagnosis 145 (75.9%) 123 (74.5%) 22 (84.6%) 0.33
Systematic diagnosis
  Cardiovascular 127 (66.5%) 111 (67.3%) 16 (61.5%)
  Respiratory 25 (13.1%) 22 (13.3%) 3 (11.5%)
  Gastrointestinal 30 (15.7%) 25 (15.2%) 5 (19.2%)
  Neurology 1 (0.5%) 1 (0.6%) 0 (0%)
   Sepsis 2 (1.0%) 1 (0.6%) 1 (3.8%) 0.48
   Hematology 1 (0.5%) 1 (0.6%) 0 (0%)
   Renal/genitourinary 2 (1.0%) 2 (1.2%) 0 (0%)
   Gynecology 1 (0.5%) 1 (0.6%) 0 (0%)
   Other 2 (1.0%) 1 (0.6%) 1 (3.8%)
Cause of mechanical ventilation
   Postoperative acute respiratory
143 (74.9%) 123 (74.5%) 20 (76.9%)
failure
   Pneumonia 16 (8.4%) 15 (9.1%) 1 (3.8%)
   Sepsis 5 (2.6%) 3 (1.8%) 2 (7.7%)
0.39
   Cardiac failure 10 (5.2%) 9 (5. 5%) 1 (3.8%)
   Coma 1 (0.5%) 1 (0.6%) 0 (0%)
   Cardiac arrest 8 (4.2%) 6 (3.6%) 2 (7.7%)
   Other 8 (4.2%) 8 (4.8%) 0 (0%)
8 (8–8) 8 (8–8) 8 (7.9–8)
ETT diameter, mm (among male) 0.51
n = 124 n = 116 n = 8
7 (7–7) 7 (7–7) 7 (7–7)
ETT diameter, mm (among female) 0.32
n = 67 n = 49 n = 18

Table 1.  Baseline characteristics according to the CLT result. Data are presented as median and interquartile
range or number (percentage). The positive CLT was defined as the cuff leak volume ≤ 110 mL and/or the
percent cuff leak ≤ 10% CLT cuff leak test, BMI body mass index, APACHE acute physiology, and chronic
health evaluation, COPD chronic obstructive pulmonary disease, ETT endotracheal tube.

Discussion
We conducted a prospective cohort study to evaluate the performance of CLT and the potential impact of pre-
extubation variables on the incidence of PES in critically ill patients who were extubated after successful SBT
under the unified ventilatory support setting. Though the patients with positive CLT results had significantly
smaller cuff leak volumes and percent cuff leaks than those with negative CLT results, the performance of the
CLT was insufficient for the accurate prediction of PES. The potential risk factors, female sex, and the larger BMI
were associated with CLT results. However, these factors were not significantly associated with the incidence of
PES. Frequent endotracheal suctioning during the 24-h period prior to extubation was independently associated
with the incidence of PES.
As the direct visualization of laryngeal edema is difficult with an endotracheal tube in position, the CLT has
been widely accepted to predict PES incidence. However, this study contributes to the limited and conflicting
evidence regarding the predictive accuracy of CLT. The standard rate of patients deemed positive through the CLT
(the cuff leak volume ≤ 110 mL) has been reported to be between 5 and 30% in critically ill adult s­ ettings11,20,23,27,
which validates our finding of 13.6%. Simultaneously, the incidence of PES in this study was 9.9%, which was
comparable to the results of previous studies with a lower rate than the positive CLT rate. Thus, the relatively
low prevalence of PES has been regarded as a possible factor for the poor performance of the CLT in the iden-
­ ccurrence3. Laryngeal edema should occur in most intubated patients
tification of patients at high risk of PES o
with varying degrees. The quantity of data on airway stenosis regarding PES and respiratory distress incidence

Scientific Reports | (2021) 11:19993 | https://doi.org/10.1038/s41598-021-99501-8 4

Vol:.(1234567890)
www.nature.com/scientificreports/

All Negative CLT Positive CLT p value

SOFA score before extubation 8 (6–10) 8 (6–10) 8 (7–9) 0.77
Duration of mechanical ventilation, h 93 (45–197) 93 (46–194) 90 (40–231) 0.49
Arterial blood gas levels and respiratory data during the successful SBT
   pH 7.43 (7.40–7.46) 7.43 (7.40–7.46) 7.43 (7.40–7.46) 0.90
   ­PaCO2, mmHg 41.4 (38.4–44.1) 41.3 (38.4–43.9) 41.8 (39.3–45.2) 0.48
   ­PaO2/FiO2, mmHg 314 (253–379) 314 (257–387) 308 (247–359) 0.61
   Rapid shallow breathing index, breaths/min/L 36 (27–50) 38 (28–50) 37 (27–48) 0.98
   ­SpO2, % 99 (97–100) 99 (97–100) 98 (97–99) 0.22
Data during 24 h before extubation
0.4 (− 1.1 to 2.1) 0.4 (− 1.2 to 2) 0.7 (− 0.3 to 2.7)
   Increased body weight from ICU admission, kg 0.28
n = 187 n = 161 n = 26
   Number of endotracheal suctioning 14 (11–17) 14 (11–17.3) 13.5 (11.3–16.5) 0.56
   Fluid balance, mL  − 365 (− 1222 to − 179)  − 365 (− 1204 to − 214)  − 94 (− 1151 to − 44) 0.71
   Systemic steroid therapy 16 (8.4%) 12 (7.3%) 4 (15.4%) 0.24
Cuff leak test
   Cuff leak volume, mL 283 (184–391) 301 (224–412) 53 (12–93)  < 0.001
   Percent cuff leak, % 57.3 (40.9–75.3) 61.9 (46.2–76.3) 11.5 (2.6–21.1)  < 0.001
   Cuff leak volume ≤ 110 mL 26 (13.6%) – 26 (100%) –
   Percent cuff leak ≤ 10% 9 (4.7%) – 9 (34.6%) –

Table 2.  Data at extubation and the detail of CLT. Data are presented as median and interquartile range or
number (percentage). The positive CLT was defined as the cuff leak volume ≤ 110 mL and/or the percent cuff
leak ≤ 10% CLT cuff leak test, SOFA sequential organ failure assessment, SBT spontaneous breathing trial.

Patients without PES (n = 172) Patients with PES (n = 19) p value

Age, years 64 (52–74) 69 (62.5–74.5) 0.34
Sex (female) 60 (34.9%) 7 (36.8%) 1.00
2 22.5 (19.9–25.4) 23.7 (21.1–27.0)
BMI, kg/m 0.28
n = 170 n = 19
54 (31.4%) 7 (36.8%)
BMI > 25 kg/m2 0.80
n = 170 n = 19
APACHE II score 15 (12–19) 14 (11.5–20) 0.83
APACHE III score 59 (48–71) 66 (49–74) 0.42
Comorbidities
   Heart failure 65 (37.8%) 7 (36.8%) 1.00
   Asthma 5 (2.9%) 2 (10.5%) 0.15
   COPD 7 (4.1%) 1 (5.3%) 0.58
   Interstitial lung disease 10 (5.8%) 2 (10.5%) 0.35
Surgical diagnosis 133 (77.3%) 12 (63.2%) 0.17
Cause of mechanical ventilation
   Postoperative acute respiratory failure 131 (76.2%) 12 (63.2%)
   Pneumonia 13 (7.6%) 3 (15.8%)
   Sepsis 3 (1.7%) 2 (10.5%)
   Cardiac failure 9 (5.2%) 1 (5.3%) 0.186
   Coma 1 (0.6%) 0 (0%)
   Cardiac arrest 7 (4.1%) 1 (5.3%)
   Other 8 (4.7%) 0 (0%)
8 (8–8) 8 (8–8)
ETT diameter, mm (among male) 0.93
n = 112 n = 12
7 (7–7) 7 (7–7.3)
ETT diameter, mm (among female) 0.33
n = 60 n = 7

Table 3.  Characteristics of patients with or without PES. Data are presented as median and interquartile range
or number (percentage). PES post-extubation stridor, BMI body mass index, APACHE acute physiology, and
chronic health evaluation, COPD chronic obstructive pulmonary disease, ETT endotracheal tube.

Scientific Reports | (2021) 11:19993 | https://doi.org/10.1038/s41598-021-99501-8 5

Vol.:(0123456789)
 www.nature.com/scientificreports/

Patients without PES Patients with PES p value

SOFA score at extubation 8 (6–10) 8 (7–10) 0.67
Duration of total mechanical ventilation, hour 90 (44–170) 194 (93–677) 0.01
Prolonged total mechanical ventilation > 4 days 80 (46.5%) 14 (73.7%) 0.03
Arterial blood gas levels and respiratory data during the successful SBT
   pH 7.43 (7.40–7.46) 7.43 (7.41–7.45) 0.64
   ­PaCO2, mmHg 41.2 (38.4–43.9) 43.2 (40.9–46.2) 0.07
   ­PaO2/FiO2, mmHg 313 (251–386) 314 (289–340) 0.61
   Rapid shallow breathing index, breaths/min/L 38 (27–50) 39 (29–46) 0.97
   ­SpO2, % 99 (97–100) 99 (98–99) 0.75
Data during 24 h before extubation
0.4 (− 1 to 2.1) 0.4 (− 1.2 to 2.0)
   Increased body weight from ICU admission, kg 0.86
n = 169 n = 18
   Number of endotracheal suctioning 14 (11–17) 17 (13.5–20) 0.04
   Endotracheal suctioning > 15 times 63 (36.6%) 13 (68.4%) 0.01
   Fluid balance, mL  − 352 (− 1129 to 90)  − 526 (− 1305 to − 203) 0.30
   Systemic steroid therapy 14 (8.1%) 2 (10.5%) 0.67
Cuff leak test
   Cuff leak volume ≤ 110 mL 23 (13.4%) 3 (15.8%) 0.73
   Percent cuff leak ≤ 10% 7 (4.1%) 2 (10.5%) 0.22
   Positive CLT 23 (13.4%) 3 (15.8%) 0.73

Table 4.  Data at extubation among the patients with or without PES. Data are presented as median and
interquartile range or number (percentage). The positive CLT was defined as the cuff leak volume ≤ 110 mL
and/or the percent cuff leak ≤ 10% PES post-extubation stridor, SOFA sequential organ failure assessment, SBT
spontaneous breathing trial, CLT cuff leak test.

All Patients without PES Patients with PES p value

Reintubation within 48 h, all cause 24 (12.6%) 14 (8.1%) 10 (52.6%)  < 0.001
Treatment of PES
   Intravenous steroid administration for post-extubation stridor 5/191 (26.2%) – 5/19 (26.3%) –
   Adrenaline inhalation for post-extubation stridor 0 (0%) – 0 (0%) –
   Reintubation for post-extubation stridor 8/191 (4.2%) – 8/19 (42.1%) –
ICU length of stay, day 8 (5–14) 8 (4.5–14) 12 (8–30.5) 0.01
Hospital length of stay, day 46 (28–80) 46 (28–80) 46 (28–81) 0.86
ICU mortality 6 (3.1%) 5 (2.9%) 1 (5.3%) 0.47
Hospital mortality 15 (7.9%) 12 (7.0%) 3 (15.8%) 0.18

Table 5.  Post-extubation outcomes of patients with and without PES. Data are presented as median and
interquartile range or number (percentage). PES post-extubation stridor, ICU intensive care unit.

Univariate analysis Multivariate analysis

Odds ratio (95% CI) p value Adjusted odds ratio* (95% CI) p value
Positive CLT 1.21 (0.33–4.5) 0.77
Endotracheal suctioning > 15 times during the 24-h period
3.71 (1.34–10.3) 0.01 2.97 (1.01–8.77) 0.048
prior to the extubation
Prolonged mechanical ventilation > 4 days 3.22 (1.11–9.33) 0.03 2.43 (0.79–7.42) 0.12

Table 6.  Univariate and multivariate logistic regression analyses for PES. PES post-extubation stridor, CI
confidence interval, CLT cuff leak test. *Adjusted by gender, BMI > 25 kg/m2, and positive CLT.

is ­limited29–31. Furthermore, after prolonged endotracheal intubation, bronchoscopy showed that the tracheal
granulation tissue or ulcerations spread widely from the vocal cord to the previous ETT balloon cuff ­site17,32.
As global guidelines have not standardized the ETT size based on the tracheal d ­ iameter33,34, our patients were
provided a similar ETT size regardless of the CLT results. Mild laryngeal edema encompassing the ETT balloon
cuff could be demonstrated by the CLT without significant clinical symptoms, leading to an insufficient inter-
pretation of the test r­ esult24. Moreover, the cumulative evidence for the predictive accuracy of CLTs was based

Scientific Reports | (2021) 11:19993 | https://doi.org/10.1038/s41598-021-99501-8 6

Vol:.(1234567890)
www.nature.com/scientificreports/

on a variety of mechanical ventilation modes and clinical settings during C ­ LT22. Following the reference report
by Miller et al.11, the assist-control ventilation mode was commonly used, just as in this s­ tudy35. However, the
fact that rigorous and reproducible CLT assessment may require firm sedation and paralysis to avoid erroneous
measurement caused by coughing or ­agitation36 and the potential influence of glossoptosis on the head position
are challenges for the current standard CLT ­method37.
Furthermore, systemic steroid therapy before elective extubation has been associated with significant PES
incidence ­reductions5,25,38,39. A recent meta-analysis including 11 trials of 2472 participants revealed that prophy-
lactic systemic steroid therapy was associated with a reduced incidence of post-extubation airway events (risk
ratio [RR], 0.43; 95% CI 0.29–0.66) and reintubation (RR, 0.42; 95% CI 0.25–0.71) when compared with placebo
or no ­treatment29,40. The authors also screened six trials that documented the adverse effects and concluded
that the short duration of systemic steroid administration rarely resulted in severe adverse effects. There was no
hyperglycemia or gastrointestinal bleeding incidence, and only one patient who received systemic steroids was
recognized with the infection. Thus, the current guidelines indicate the use of prophylactic systemic steroids in
patients with airway ­compromise12,41. In our cohort, patients received systemic steroid therapy for at least 4 h
before extubation for all causes (7.3% and 15.4% of the patients with negative and positive CLT results, respec-
tively). Systemic steroid therapy was not individually associated with PES incidence (p = 0.67); however, it may
potentially treat laryngeal edema and influence patient outcomes.
Multiple investigations of the risk factors for the incidence of PES have been reported, with conflicting conclu-
sions. In this study, female sex and larger BMI were associated with CLT results. Patients with PES had a longer
duration of intubation and frequent endotracheal suctioning than those without PES. However, multivariate
analysis revealed that frequent endotracheal suctioning during the 24-h period before extubation was indepen-
dently associated with PES incidence in our patient population. The number of endotracheal suctioning indicates
the amount of trachea-bronchial secretion. Excessive tracheal-bronchial secretions can lead to bronchial plugging
and atelectasis, which can result in respiratory failure. Recent reports substantiated the significance of excessive
tracheal-bronchial secretions before extubation as risk factors for ­reintubation8,42–44. In addition, the tracheal-
bronchial secretions are relevant indicators of laryngeal edema because they are triggered simultaneously, and
both are caused by airway ­inflammation45. Furthermore, acute venous congestion due to excessive fluid infusion
could increase pulmonary congestion causing tracheal-bronchial s­ ecretions46,47; moreover, it could also increase
the risk of laryngeal ­edema48. Subsequently, the procedure of endotracheal suctioning itself has been reported
to potentially cause tracheal-bronchial i­nflammation49,50 and pulmonary e­ dema51. Thus, laryngeal edema and
tracheal-bronchial secretions are caused in conjunction with each other, however, their relative importance
has not been sufficiently investigated. The only previous study to research the association between the number
of endotracheal suctioning and laryngeal injury was in 1981, in a single prospective cohort of 150 critically ill
patients who received mechanical ventilation with endotracheal intubation or ­tracheostomy2. Stauffer et al.
reported that endotracheal suctioning frequency was not significantly associated with total laryngeal injury at
autopsy. However, this analysis included 51 patients (34%) with a tracheostomy, and the liberation process was
not standardized. According to the current clinical practice and the unified weaning readiness techniques, the
present data provide a reasonable impact.
Our findings imply that a positive result of CLT is relatively common in critically ill patients who have under-
gone invasive mechanical ventilation for more than 24 h. However, they also imply that the CLT results are not
significantly associated with PES incidence, which results in an increased reintubation rate and ICU length of
stay. Despite the lack of a significant impact on demographic variables, we found that frequent endotracheal
suctioning was an independent risk factor for PES incidence. This finding provides a rationale for further inves-
tigations of the risk management of upper airway obstruction.
Our study has several strengths. Relatively few studies have assessed the physiological or descriptive data at
extubation as risk factors for PES incidence. Though the population we investigated was relatively small, our
prospective cohort was representative, as the rate of positive results of CLT and the incidence of PES were com-
parable to those described in other recent studies. In addition, the rigorous SBT and CLT assessment revealed a
robust evaluation of CLT and risk factors for the prediction of PES.
Some limitations of this study should be acknowledged. First, it is a single-center study, with all the limita-
tions inherent in such a design. The results may not be generalizable to all geographical regions. Second, we
employed the SBT method with pressure support ventilation (PSV) using low PS and PEEP levels, following the
international guideline that recommends inspiratory pressure a­ ugmentation52. General SBT methods, both PSV
and T-piece, were reported to have comparable predictive performance for successful ­extubation53. However,
low level of assistance without PEEP has recently been recommended as a SBT m ­ ethod54, which is distinct from
the method used in this study. Third, based on previous studies reporting PES rates of 4–10%, we conducted
a 2-year observational study to obtain at least 10 patients presenting with PES in an ICU with approximately
50–150 eligible patients per year to clarify its predictive performance using binary logistic regression analysis. To
detect the association between the CLT results and PES incidence with 95% certainty at 80% power, we assumed
that a total of 184 patients would be required. This was based on CLT positivity and PES incidence rates of 18%
and 4%, respectively, according the results of a previous study that included the largest number of ­patients27. We
found that that 19 among 191 study patients had PES and could accumulate sufficient cases for analysis, however,
the CLT positivity rate and the incidence of PES in this study suggested that a greater number of patients were
warranted to determine the predictive ability of CLT for PES. Hence, our results should be examined in a meta-
analysis and in a pooled analysis. In addition, this study included all patients with mechanical ventilation and
not exclusively patients at high risk for PES, for which the guideline recommends ­CLT12. Furthermore, patients
with unplanned extubation regarded as high risk for P ­ ES12 were not included in the study due to lack of SBT or
CLT, and we did not identify any patient who was reintubated after unplanned extubation and then underwent
planned extubation. Forth, the formal PES diagnosis was based on clinical symptoms and, therefore, relatively

Scientific Reports | (2021) 11:19993 | https://doi.org/10.1038/s41598-021-99501-8 7

Vol.:(0123456789)
 www.nature.com/scientificreports/

subjective. Fifth, although the assessment of SBT and CLT was standardized, the final decision of extubation, the
administration of systemic steroids, the treatment of patients with PES, and the decision of reintubation were at
the discretion of the clinician in charge of the patients without specific institutional protocols.

Conclusions
In the critically ill patients receiving mechanical ventilation, patients who required frequent endotracheal suc-
tioning were at a high risk of developing PES. Though its impact on the incidence of PES and patient outcomes
needs to be further explored, frequent endotracheal suctioning before extubation could be added to the usual
prediction for PES.

Data availability
The datasets generated during and/or analyzed during the current study are available from the corresponding
author on reasonable request.

Received: 20 March 2021; Accepted: 24 September 2021

References
1. Colice, G. L., Stukel, T. A. & Dain, B. Laryngeal complications of prolonged intubation. Chest 96, 877–884 (1989).
2. Stauffer, J. L., Olson, D. E. & Petty, T. L. Complications and consequences of endotracheal intubation and tracheotomy. A prospec-
tive study of 150 critically ill adult patients. Am. J. Med. 70, 65–76 (1981).
3. Pluijms, W. A., van Mook, W. N., Wittekamp, B. H. & Bergmans, D. C. Postextubation laryngeal edema and stridor resulting in
respiratory failure in critically ill adult patients: Updated review. Crit. Care 19, 295 (2015).
4. Patel, A. B., Ani, C. & Feeney, C. Cuff leak test and laryngeal survey for predicting post-extubation stridor. Indian J. Anaesth. 59,
96–102 (2015).
5. Cheng, K. C. et al. Methylprednisolone reduces the rates of postextubation stridor and reintubation associated with attenuated
cytokine responses in critically ill patients. Minerva Anestesiol. 77, 503–509 (2011).
6. Jaber, S. et al. Post-extubation stridor in intensive care unit patients. Risk factors evaluation and importance of the cuff-leak test.
Intensive Care Med. 29, 69–74 (2003).
7. Schnell, D. et al. Cuff leak test for the diagnosis of post-extubation stridor: a multicenter evaluation study. J. Intensive Care Med.
34, 391–396 (2019).
8. Epstein, S. K. & Ciubotaru, R. L. Independent effects of etiology of failure and time to reintubation on outcome for patients failing
extubation. Am. J. Respir. Crit. Care Med. 158, 489–493 (1998).
9. Frutos-Vivar, F. et al. Outcome of reintubated patients after scheduled extubation. J. Crit. Care 26, 502–509 (2011).
10. Seymour, C. W., Martinez, A., Christie, J. D. & Fuchs, B. D. The outcome of extubation failure in a community hospital intensive
care unit: A cohort study. Crit. Care 8, R322–R327 (2004).
11. Miller, R. L. & Cole, R. P. Association between reduced cuff leak volume and postextubation stridor. Chest 110, 1035–1040 (1996).
12. Girard, T. D. et al. An Official American Thoracic Society/American College of Chest physicians clinical practice guideline: Lib-
eration from mechanical ventilation in critically Ill adults. Rehabilitation protocols, ventilator liberation protocols, and cuff leak
tests. Am. J. Respir. Crit. Care Med. 195, 120–133 (2017).
13. Baptistella, A. R. et al. Prediction of extubation outcome in mechanically ventilated patients: Development and validation of the
extubation predictive score (ExPreS). PLoS ONE 16, e0248868 (2021).
14. Lewis, K. et al. The cuff leak test in critically ill patients: An international survey of intensivists. Acta Anaesthesiol. Scand. 65,
1087–1094 (2021).
15. Moran, J. V., Godil, S. A., Goldner, B., Godil, K. & Aslam, J. Post-extubation stridor complicating COVID-19-associated acute
respiratory distress syndrome: A case series. Cureus 12, e10492 (2020).
16. Wang, C. L. et al. The role of the cuff leak test in predicting the effects of corticosteroid treatment on postextubation stridor. Chang.
Gung. Med. J. 30, 53–61 (2007).
17. Chung, Y. H., Chao, T. Y., Chiu, C. T. & Lin, M. C. The cuff-leak test is a simple tool to verify severe laryngeal edema in patients
undergoing long-term mechanical ventilation. Crit. Care Med. 34, 409–414 (2006).
18. Keeratichananont, W., Limthong, T. & Keeratichananont, S. Cuff leak volume as a clinical predictor for identifying post-extubation
stridor. J. Med. Assoc. Thai. 95, 752–756 (2012).
19. Sandhu, R. S., Pasquale, M. D., Miller, K. & Wasser, T. E. Measurement of endotracheal tube cuff leak to predict postextubation
stridor and need for reintubation. J. Am. Coll. Surg. 190, 682–687 (2000).
20. Zhou, T. et al. Cuff-leak test for predicting postextubation airway complications: A systematic review. J. Evid. Based Med. 4, 242–254
(2011).
21. Ochoa, M. E. et al. Cuff-leak test for the diagnosis of upper airway obstruction in adults: A systematic review and meta-analysis.
Intensive Care Med. 35, 11711179 (2009).
22. Kuriyama, A., Jackson, J. L. & Kamei, J. Performance of the cuff leak test in adults in predicting post-extubation airway complica-
tions: A systematic review and meta-analysis. Crit. Care 24, 640 (2020).
23. Shin, S. H. et al. The cuff leak test is not predictive of successful extubation. Am. Surg. 74, 1182–1185 (2008).
24. Gros, A. et al. Intra-individual variation of the cuff-leak test as a predictor of post-extubation stridor. Respir. Care 57, 2026–2031
(2012).
25. Ho, L. I., Harn, H. J., Lien, T. C., Hu, P. Y. & Wang, J. H. Postextubation laryngeal edema in adults. Risk factor evaluation and
prevention by hydrocortisone. Intensive Care Med. 22, 933–936 (1996).
26. Erginel, S., Ucgun, I., Yildirim, H., Metintas, M. & Parspour, S. High body mass index and long duration of intubation increase
post-extubation stridor in patients with mechanical ventilation. Tohoku J. Exp. Med. 207, 125–132 (2005).
27. Kriner, E. J., Shafazand, S. & Colice, G. L. The endotracheal tube cuff-leak test as a predictor for postextubation stridor. Respir.
Care 50, 1632–1638 (2005).
28. Ferreira, F. L., Bota, D. P., Bross, A., Melot, C. & Vincent, J. L. Serial evaluation of the SOFA score to predict outcome in critically
ill patients. JAMA 286, 1754–1758 (2001).
29. Wittekamp, B. H., van Mook, W. N., Tjan, D. H., Zwaveling, J. H. & Bergmans, D. C. Clinical review: Post-extubation laryngeal
edema and extubation failure in critically ill adult patients. Crit. Care 13, 233 (2009).
30. Mackle, T., Meaney, J. & Timon, C. Tracheoesophageal compression associated with substernal goitre. Correlation of symptoms
with cross-sectional imaging findings. J. Laryngol. Otol. 121, 358–361 (2007).
31. Ding, L. W., Wang, H. C., Wu, H. D., Chang, C. J. & Yang, P. C. Laryngeal ultrasound: A useful method in predicting post-extubation
stridor. A pilot study. Eur. Respir. J. 27, 384–389 (2006).

Scientific Reports | (2021) 11:19993 | https://doi.org/10.1038/s41598-021-99501-8 8

Vol:.(1234567890)
www.nature.com/scientificreports/

32. Kastanos, N., Estopa Miro, R., Marin Perez, A., Xaubet Mir, A. & Agusti-Vidal, A. Laryngotracheal injury due to endotracheal
intubation: Incidence, evolution, and predisposing factors. A prospective long-term study. Crit. Care Med. 11, 362–365 (1983).
33. Higgs, A. et al. Guidelines for the management of tracheal intubation in critically ill adults. Br. J. Anaesth. 120, 323–352 (2018).
34. Quintard, H. et al. Experts’ guidelines of intubation and extubation of the ICU patient of French Society of Anaesthesia and Inten-
sive Care Medicine (SFAR) and French-speaking Intensive Care Society (SRLF): In collaboration with the pediatric Association of
French-speaking Anaesthetists and Intensivists (ADARPEF), French-speaking Group of Intensive Care and Paediatric emergencies
(GFRUP) and intensive care physiotherapy society (SKR). Ann. Intensive Care 9, 13 (2019).
35. Lewis, K. et al. Cuff leak test and airway obstruction in mechanically ventilated ICU patients: A pilot randomised controlled trial.
Ann. Am. Thorac. Soc. https://​doi.​org/​10.​1513/​annal​sats.​202103-​390oc (2021).
36. Pettignano, R., Holloway, S. E., Hyman, D. & LaBuz, M. Is the leak test reproducible? S. Med. J. 93, 683–685 (2000).
37. Goto, S., Ishikawa, J. Y., Idei, M. & Nomura, T. False-positive cuff leak test due to glossoptosis. Am. J. Respir. Crit. Care Med. https://​
doi.​org/​10.​1164/​rccm.​202104-​0820IM (2021).
38. Darmon, J. Y. et al. Evaluation of risk factors for laryngeal edema after tracheal extubation in adults and its prevention by dexa-
methasone. A placebo-controlled, double-blind, multicenter study. Anesthesiology 77, 245–251 (1992).
39. François, B. et al. 12-h pretreatment with methylprednisolone versus placebo for prevention of postextubation laryngeal oedema:
A randomised double-blind trial. The Lancet 369, 1083–1089 (2007).
40. Kuriyama, A., Umakoshi, N. & Sun, R. Prophylactic corticosteroids for prevention of postextubation stridor and reintubation in
adults: A systematic review and meta-analysis. Chest 151, 1002–1010 (2017).
41. Popat, M. et al. Difficult Airway Society Guidelines for the management of tracheal extubation. Anaesthesia 67, 318–340 (2012).
42. Salam, A., Tilluckdharry, L., Amoateng-Adjepong, Y. & Manthous, C. A. Neurologic status, cough, secretions and extubation
outcomes. Intensive Care Med. 30, 1334–1339 (2004).
43. Mokhlesi, B. et al. Predicting extubation failure after successful completion of a spontaneous breathing trial. Respir. Care 52,
1710–1717 (2007).
44. Khamiees, M., Raju, P., DeGirolamo, A., Amoateng-Adjepong, Y. & Manthous, C. A. Predictors of extubation outcome in patients
who have successfully completed a spontaneous breathing trial. Chest 120, 1262–1270 (2001).
45. Rogers, D. F. Physiology of airway mucus secretion and pathophysiology of hypersecretion. Respir. Care 52, 1134–1146 (2007).
46. Ngo, C., Dahlmanns, S., Vollmer, T., Misgeld, B. & Leonhardt, S. Object-oriented modeling of thoracic fluid balance to study
cardiogenic pulmonary congestion in humans. Comput. Methods Progr. Biomed. 180, 104998 (2019).
47. Solymosi, E. A. et al. Chloride transport-driven alveolar fluid secretion is a major contributor to cardiogenic lung edema. Proc.
Natl. Acad. Sci. U.S.A. 110, E2308–E2316 (2013).
48. Jan, R., Alahdal, A. & Bithal, P. K. Effect of two regimens of fluid administration on airway edema in prone-position surgery.
Anesth. Essays Res. 14, 467–473 (2020).
49. Sole, M. L., Bennett, M. & Ashworth, S. Clinical indicators for endotracheal suctioning in adult patients receiving mechanical
ventilation. Am. J. Crit. Care 24, 318–324 (2015).
50. Sackner, M. A., Landa, J. F., Greeneltch, N. & Robinson, M. J. Pathogenesis and prevention of tracheobronchial damage with suc-
tion procedures. Chest 64, 284–290 (1973).
51. Bhattacharya, M., Kallet, R. H., Ware, L. B. & Matthay, M. A. Negative-pressure pulmonary edema. Chest 150, 927–933 (2016).
52. Ouellette, D. R. et al. Liberation from mechanical ventilation: An official American College of Chest Physicians/American Thoracic
Society Clinical Practice Guideline: Inspiratory pressure augmentation during spontaneous breathing trials, protocols minimizing
sedation, and non-invasive ventilation immediately after extubation. Chest 151, 166–180 (2017).
53. Li, Y., Li, H. & Zhang, D. Comparison of T-piece and pressure support ventilation as spontaneous breathing trials in critically ill
patients: A systematic review and meta-analysis. Crit. Care 24, 67 (2020).
54. Jung, B., Vaschetto, R. & Jaber, S. T. tips to optimize weaning and extubation success in the critically ill. Intensive Care Med. 46,
2461–2463 (2020).

Acknowledgements
We would like to thank all the staff of the participating ICU for the supervision of this project and the assistance
with the data collection.

Author contributions
Study concept and design: A.T., A.U. and Y.F. Data acquisition: A.T., Y.H., R.H., R.S., Y.K., H.E., T.Y., A.M., K.S.,
D.H., N.I. and N.O. Data processing and interpretation: A.T., Y.H., R.H., R.S., Y.K., H.E., T.Y., A.M., K.S., D.H.,
N.I. and N.O. Statistical analysis: A.T., A.U. and Y.F. Figure preparation: A.T. Drafting of the manuscript: A.T.
Revision of the manuscript for important intellectual content: A.T., A.U., T.Y. and Y.F. All authors reviewed the
manuscript.

Competing interests 
The authors declare no competing interests.

Additional information
Supplementary Information The online version contains supplementary material available at https://​doi.​org/​
10.​1038/​s41598-​021-​99501-8.
Correspondence and requests for materials should be addressed to A.T.
Reprints and permissions information is available at www.nature.com/reprints.
Publisher’s note  Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.

Scientific Reports | (2021) 11:19993 | https://doi.org/10.1038/s41598-021-99501-8 9

Vol.:(0123456789)
 www.nature.com/scientificreports/

Open Access  This article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the
Creative Commons licence, and indicate if changes were made. The images or other third party material in this
article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from
the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/.

© The Author(s) 2021

Scientific Reports | (2021) 11:19993 | https://doi.org/10.1038/s41598-021-99501-8 10

Vol:.(1234567890)