Association of extubation failure rate with patient outcomes

Ivan Gur; Roei Tounek; Ronen Zalts; Rona Epshtein; Asaf Miller; Amichai Gutgold

doi:10.4266/acc.004250

Articles

Page Path: HOME > Acute Crit Care > Volume 41(2); 2026 > Article

Original Article Pulmonary Association of extubation failure rate with patient outcomes: Acute and Critical Care 2026;41(2):295-303.
DOI: https://doi.org/10.4266/acc.004250
Published online: April 17, 2026

Ivan Gur^1,2

, Roei Tounek³

, Ronen Zalts^1,2

, Rona Epshtein⁴, Asaf Miller², Amichai Gutgold²

¹Department of Internal Medicine C, Rambam Health Care Campus, Haifa, Israel

²Medical Intensive Care Unit, Rambam Health Care Campus, Haifa, Israel

³The Ruth and Bruce Rappaport Faculty of Medicine, Technion Israel Institute of Technology, Haifa, Israel

⁴Faculty of Medicine, the Hebrew University of Jerusalem, Jerusalem, Israel

Corresponding author: Ivan Gur Department of Internal Medicine C, Rambam Health Care Campus, 8 HaAlia St, Haifa 3109601, Israel. email: I_GUR@rambam.health.gov.il

• Received: September 22, 2025 • Revised: January 2, 2026 • Accepted: January 8, 2026

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

617 Views
20 Download
1 Scopus

prev next

Full Article

Download PDF

Abstract
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
KEY MESSAGES
NOTES
SUPPLEMENTARY MATERIALS
REFERENCES

Abstract

Background
While the propensity of evidence indicates the potential harm of failed extubation attempts, avoidance of any such risk may unnecessarily expose patients to prolonged invasive ventilation. We aimed to study the effects of extubation failure rate (EFR) on patient mortality and ventilation-free days (VFD).
Methods
Adult patients admitted to the medical intensive care unit who underwent planned extubation were included. Extubation failure was defined as death or return to invasive positive pressure ventilation within 7 days from extubation. The primary outcome was 30-day mortality and the secondary outcome was VFD. For each calendar month, the average 30-day mortality or VFD was plotted against the EFR, with polynomial regression models of increasing complexity fitted until no further increase in the adjusted R² could be achieved.
Results
Of the 774 patients included in final analysis, 262 (33.8%) failed extubation. Matched by the propensity for extubation failure, the 30-day mortality analysis for both groups showed no significant difference (18.4% vs. 16.2%, P=0.34). The relationship between monthly EFR and 30-day survival or VFD was best described by a quadratic regression model (adjusted R²=0.816 and 0.624, respectively). Based on this model, the optimal EFR was calculated at 33.1% (for 30-day survival) and 28.8% (for VFD).
Conclusions
Our data support the notion of an optimal EFR. This optimum may be higher than the minimal failure rate achievable.
Key Words: airway extubation; prognosis; treatment failure; ventilator weaning

INTRODUCTION

Extubation failure, defined as the need for intubation within 2–7 days after planned extubation [1,2], occurs in around 10%–40% of intensive care unit (ICU) patients who are mechanically ventilated [3]. Risk factors for extubation failure have been extensively studied, with various predictive models and decision assistance scores devised to help prognosticate the chances of unsuccessful extubation and the need to reintubate [4]. Extubation failure is a common primary outcome in studies examining interventions aimed at improving weaning from mechanical ventilation and decreasing potential harms of mechanical ventilation [5-7].

Conversely, the harms of invasive positive pressure ventilation are well studied and widely accepted [8,9]. Since prediction models for successful weaning are far from perfect [10,11], lowering the extubation failure rate (EFR) will doubtlessly come at the cost of prolonging invasive ventilation in at least some patients. A paucity of evidence exists as to the effects of EFR on the entire mechanically ventilated patient population. This notion was demonstrated by Kapnadak et al. [12] when comparing the average monthly ventilation-free days (VFD) of patients extubated during months with low, intermediate, and high EFR. Average VFD was lower in the middle tertile in terms of EFR, compared to both the high and low EFR. Mortality was not reported in their study. Although several other field experts have suggested the optimal EFR to be 5%–15% [3,13], this, to the best of our knowledge, is not based on experimental data beyond Kapnadak's work [12]. In this study, we aimed to explore the association of EFR with patient mortality and VFD.

MATERIALS AND METHODS

This study was conducted according to the Declaration of Helsinki. This study was approved by the Ethics Committee of Rambam Health Care (No. RMB-22-0249). The requirement for consent was waived by the Ethics Committee due to the purely observational nature of this study.

Design and Setting

In this retrospective cohort study we examined the individual data of all adult patients admitted between January 1, 2016 and December 31, 2022 to the 12-bed medical ICU of a large tertiary referral center serving a population of 2 million residents. Data collected included demographics, prior diagnoses, and the reason for intubation. Vital signs and laboratory results at the time of intubation and of extubation were used to calculate the Sequential Organ Failure Assessment (SOFA) scores.

Population

The study population included all adult patients who were intubated and mechanically ventilated for at least 12 consecutive hours through an oral endotracheal tube (patients ventilated through tracheostomy or using noninvasive positive pressure ventilation were not included). Patients were excluded if: (1) they were younger than 18 years of age; (2) extubation was not planned (i.e., accidental or self extubation); (3) a planned extubation was not attempted for any reason; or (4) a standing do-not-reintubate order was recorded.

Extubation and Reintubation

Typically, before attempted extubation, patients were placed on a spontaneous breathing trial consisting of low (5–10 cm H₂O) pressure support on top of 5–8 cm H₂O positive end-expiratory pressure. Fractional inspired oxygen was usually 30%–50%. A positive leak test, hemodynamic stability, and clinical resolution or significant improvement of the condition for which intubation was indicated were prerequisites for all planned extubations. Extubation failure was defined as death or return to invasive positive pressure ventilation within 7 days from extubation [14,15].

Selection of Optimal Time Unit for Analysis

We chose a time unit cutoff of 1 month for our analysis based on cross-validation results that highlighted it as the optimal balance between data granularity and interpretability. Initial cross-validation involved testing multiple time cutoffs, including weekly, quarterly, and semester-based intervals, and comparing predictive performance and model stability. In line with previously published works [12], we found the 1-month interval to be optimal, capturing sufficient temporal variation without overfitting or underfitting the model. Weekly cutoffs, while providing more data points, introduced noise and reduced the variability of EFRs (as a direct result of the low number of extubations undertaken each week). Conversely, longer intervals, such as quarterly or semester cutoffs, averaged out significant changes and critically decreased the number data points available for analysis.

Outcomes

The primary outcome was defined as mortality within 30 days from planned extubation. The secondary outcome was VFD, defined as the total number of days free from invasive ventilation within the initial 28 days from extubation [12]. Noninvasive positive pressure ventilatory support, including high flow, constant, and bilevel positive pressure devices, were not considered invasive ventilation and were thus included in VFD counts for the purpose of this analysis. On the other hand, if a patient failed the weaning attempt and had to undergo tracheostomy, periods without continuous ventilatory support were not included in the VFD count.

Statistical Analysis

Demographic and clinical data, including background diagnoses, the reason for ventilatory support, the major underlying pathophysiological mechanism requiring intensive care, and vital signs and laboratory results upon intubation and extubation were analyzed using descriptive statistics, parametric, and nonparametric tests as appropriate. To accurately assess the propensity for extubation failure, we constructed a univariate regression model for all clinical data (including vitals and laboratories at the time of intubation and at the time of extubation) for both extubation failure (a univariate logistic regression) and the primary outcome of 30-day mortality (Cox’s regression). Only variables found to significantly predict both, provided no significant inter-correlation was observed (i.e., Pearson’s r<0.7), were included in a multivariate logistic regression model based propensity scoring. The resultant propensity score was used to match patients who failed the extubation attempt at a 1:1 ratio to successfully weaned patients using an optimal approach without replacement with a caliper size of <0.2 standard deviations. To model the relationship between EFR and the primary or secondary outcomes, we evaluated polynomial regression models of increasing complexity until no further improvement in the adjusted R² was noted. That is, we examined linear, quadratic, cuboid, fourth-order and fifth-order regression models, calculating R² for each, and chose the one with the optimal R². All analyses were performed using R version 4.4.1 (R Foundation for Statistical Computing).

RESULTS

During the study period, 2,813 patients were admitted to the medical ICU, of whom 1,342 (47.7%) required invasive mechanical ventilatory support via an oral endotracheal tube. After excluding four patients due to early transfer or incomplete documentation, 778 patients (57.9%) underwent planned extubation, and were thus included in the final analysis. The remainder had no attempted extubation, resulting in either the patient’s demise or tracheostomy. This patient selection process is summarized in Figure 1.

Patient age was significantly higher in the extubation failure group (mean difference [MD], 7.4; 95% CI, 4.9–9.8; P<0.001). No other clinically or statistically significant differences were noted between the successful weaning and the failed extubation groups in terms of sex, the main indication for intubation, or medical history, as presented in Table 1.

Clinically, patients who failed extubation displayed more deranged physiology—at both the time of intubation, and even more so at the time of extubation. SOFA scores were higher in the failed extubation group at both the time of intubation (MD, 0.6 points; 95% 95% CI, 0.25,0.94; P=0.0005) and extubation (MD, 1.4 points; 95% CI, 1.06–1.74; P<0.0001). Mean arterial pressure was lower in the failed extubation group at the time of extubation (MD, –5.5 mm Hg; 95% CI, –9.4 to –1.8; P=0.004) but not at the time of intubation (MD, 3.3; 95% CI, –13 to 19; P=0.59). Heart rate was higher in the failed extubation group at the time of extubation (MD, 14.7/min; 95% CI, 11.5 to 18; P<0.001) but not at the time of intubation (MD, 3.1/min; 95% CI, –0.3 to 6.5; P=0.073). Serum creatinine levels were slightly lower in the extubation failure group at the time of extubation (MD, –0.15 mg/dl; 95% CI, –0.38 to –0.08; P=0.044) but not at the time of intubation (MD, 0.001; 95% CI, –0.93 to 0.94; P=0.996). No statistically or clinically significant differences were noted between the failed extubation and successful weaning groups in terms of other vital signs or laboratory results, either at the time of initial intubation or at the time of extubation.

The propensity for extubation failure was calculated using a multivariate logistic regression model including the patient’s age and SOFA score at the time of planned extubation. After matching both groups at a 1:1 ratio, the aforementioned clinical differences were rendered statistically nonsignificant. These results are summarized in Table 2.

Survival analysis, focused on the main outcome of mortality within 30 days from planned extubation, showed a similar trend. While significantly lower for the failed extubation group in the unmatched cohort (P=0.002), there was no statistically significant mortality difference at 30 days when propensity score matching was used (P=0.34). These results are presented in Figure 2.

Analyzing the average 30-day mortality as a function of EFR for each calendar month, a clear inversely U-shaped pattern was evident, as shown in Figure 3. When fitting regression models, the first-order model demonstrated a poor fit, with an adjusted R² of 0.012. The optimal adjusted R² of 0.816 was achieved with a second-order (quadratic) model, while a third-order (cubic) model did not improve the adjusted R² (0.809) any further. Based on the quadratic regression model, the optimal EFR was estimated to be 33.1%. Of note, 26.2% of months sampled for this study had an EFR of 33.1%±5%. There seemed to be no significant difference in the EFR (33% vs. 36%, P=0.53) or average monthly 30-day survival (28.7 vs. 29.4, P=0.26) when comparing the winter months (December, January, and February) to the rest of the year. Similarly, comparing the EFR of months marked by the coronavirus disease 2019 (COVID-19) pandemic (April 2020 to December 2022) to the rest of the cohort, the differences in EFR (30.6% vs. 35.9%, P=0.63) or average monthly 30-day survival (27.6 vs. 29.5, P=0.86) were nonsignificant.

Similar trends were noted when looking at the secondary outcome of VFD. The adjusted R² was 0.002, 0.624 and 0.609 for the first-, second- and third-order models, respectively. Choosing the quadratic model, the optimal EFR was estimated at 28.8% (Figure 4). Sensitivity analysis limiting the definition of extubation failure to 72 hours excluded 65 extubation failures that occurred at an interval of 72–168 hours from extubation representing 24.8% of patients who failed extubation. Mortality and VFD trends were very similar, with an optimal EFR of 32.9% and 26.1% for 30-day mortality and VFD respectively, as presented in Supplementary Tables 1 and 2.

DISCUSSION

In this single-center retrospective study we aimed to explore the possibility of an optimal EFR, in a cohort of mechanically ventilated patients admitted to a medical ICU. We found the association between EFR and patient-centered outcomes such as VFD and all-cause mortality to be best expressed by a quadratic regression model, i.e., a U-shaped curve, with an optimal EFR for VFD and mortality of 28.8% and 33.1% respectively.

As the risk of extubation failure is dependent on patient characteristics including the indication to ventilate, different causes of ventilatory failure, disease severity, and patient comorbidities, optimal EFR is expected to be vary for each ICU’s unique population [14,16]. Yet, this study suggests that EFR is a measurable parameter that can serve as a quality index for a single ICU and may aid direct the unit towards ameliorating patient and system centered outcomes, such as decreasing overall need for ventilation and length of stay. Furthermore, a systematic review of optimal EFR could inform clinician risk tolerance, as the results could prove to be much higher than what is currently accepted. Similarly, optimal EFR could serve as a surrogate marker assessing the performance of various weaning interventions, including protocols and novel decision aids.

The current perception of extubation failure as a significant negative prognostic factor stems from the well-known adverse events associated with intubation itself, such as hemodynamic and respiratory instability, and their association with increased mortality [17]. The risk of ventilator-associated-pneumonia was shown to increase after re-intubation, but also increases with prolonged mechanical ventilation and tracheostomy [18,19]. Correspondingly, several studies demonstrated an association between extubation failure and increased mortality. [1,2]

However, this hardly establishes causality, as increased disease severity could explain both unsuccessful weaning and the decimal outcome associated with patients in whom extubation was not attempted. Some studies demonstrating an increase in ICU stay and short-term mortality lacked adequate controls (and in particular, adjustment for many important observable confounders) and had notably small sample sizes when considering the failed extubation portion of the cohort [20,21]. Others matched controls based on pathophysiology at ICU admission rather than at the point of extubation, which is very different from our data set. This methodological caveat is typical of all studies included in the most recent meta-analysis on the subject [22]. When we used propensity score matching to account for differences in physiological derangement between the successfully weaned and extubation failure groups at the time of extubation rather than at intubation or ICU admission, we found no difference in survival or VFD. This finding is in line with a previous study that found no difference in mortality when extubation failure was due to upper airway obstruction, perhaps because in this specific etiological group variability in other observable confounders (such as the severity of physiological derangement) was negligible [20]. This may also explain why in a large multicenter trial, interventions found to significantly reduce the EFR failed to decrease mortality or length of ICU stay [7]. Notably, a recent multi-center randomized trial compared 4 weaning readiness screening strategies on EFR, and found not only no association with increased mortality, but also a decrease in ICU and overall lengths of stay when the EFR increased as a result of a more permissive screening protocol [23].

To the best of our knowledge this study is the first to identify a discrete optimal EFR per specific ICU. The hypothesis of optimal EFR existence was first suggested by Macintyre et al. [13], who estimated it to be 10%–15%, and then by Krinsley et al. [3], who theorized it should be around 5-10% and claimed that the median EFR of 18% found in a literature review was due to suboptimal pre-extubation assessment. Kapnadak et al. [12] tested this by analyzing the association of EFR per month with VFD and mortality. That study found that in 1,579 extubations and 150 failed planned extubations, an EFR of 7%–15% resulted in fewer VFD then either <7% or >15%. Although their aim and general methodology resembles ours, there are some important differences. First, they compared tertiles and could identify neither a singular optimal EFR, nor a mortality difference. Second, their cohort was based on a registry used to evaluate the impact of a quality improvement program that included protocolized weaning, which could limit the external validity of these findings.

We defined extubation failure based on a 7-day window. Other studies and guidelines used other definitions, most commonly 48 or 72 hours from extubation, in an attempt to avoid including a new respiratory failure event unrelated to the index extubation [12,13,24,25]. This perception was supported by analysis of a database containing 98,267 admissions from 2000 to 2009, in which 90% of extubation failures occurred in the first 96 hours after extubation [26]. More recently however, new studies questioned the use of these definitions, as the increasing use of non-invasive ventilation post-extubation may prolong the time to re-intubation [15,23,27].

This study has several important limitations. First, inherent to the retrospective and pragmatic design, there are possible biases, mainly that the difference in outcomes may be due to a different attending intensivist, seasonal morbidity or epidemics, though we found no significant difference when comparing calendar winter and summer months, or the COVID-19 epidemic era to other non-epidemic months. Of note, in our ICU, attending physicians rotate or work in parallel throughout the month, such that differences in the attending intensivist management habits were unlikely to influence the results. Nonetheless, the retrospective design makes it virtually impossible to infer any causality in a sample of this size.

Second, an EFR of 33.7% is at the higher end of those previously reported (usually 10%–35%) [3,24,27], limiting the external validity of our findings to units with significantly lower EFR. These relatively high rates may be an expression of worse patient baseline status, our choice of extubation failure definition or lower physician risk aversion. It should be noted that the mortality we report is similar to that predicted by SOFA scores [28].

Our data support the notion of an optimal EFR. This optimum may be higher than the minimal failure rate achievable, suggesting the importance of a focus on optimizing meaningful outcomes (e.g., patient survival) rather than EFR. As a surrogate quality marker, EFR should be defined individually for each specific ICU considering the unique patient characteristics of the unit in question. Larger multicenter studies are needed to elucidate interdepartmental variability and the universal existence of an optimal EFR.

KEY MESSAGES

▪ In this retrospective analysis of 774 mechanically ventilated medical intensive care unit patients, the relationship between monthly extubation failure rate (EFR) and 30 days survival or ventilation-free days (VFD) was best described by a quadratic regression model (adjusted R²=0.816 and 0.624, respectively).

▪ This suggests very low EFRs could be harmful for the critically ill patient population.

▪ Based on this model, the optimal EFR was calculated at 33.1% (for 30 days survival) and 28.8% (for VFD).

NOTES

CONFLICT OF INTEREST

No potential conflict of interest relevant to this article was reported.

FUNDING

None.

ACKNOWLEDGMENTS

We are grateful to Dr. Yuval Nov for his illuminating suggestions in the approach to the time unit selection problem at the core of this project. We also wish to thank Mr. Tomer Karny for his invaluable assistance in retrieving the data used in this research. Data mining was supported by the “Insight” team and the MDClone interface.

AUTHOR CONTRIBUTIONS

Conceptualization: IG, AM, AG. Methodology: IG, RZ, AG. Formal analysis: IG,. Data curation: IG, RT, RE. Visualization: AM. Project administration: AM, AG. Funding acquisition: IG,. Writing - original draft: IG, AM, AG. Writing - review & editing: IG, RT, RZ, AM, AG. All authors read and agreed to the published version of the manuscript.

SUPPLEMENTARY MATERIALS

Supplementary materials can be found via https://doi.org/10.4266/acc.004250.

Supplementary Table 1.

Study patients characteristics at a 72 hours of extubation failure

acc-004250-Supplementary-Table-1.pdf

Supplementary Table 2.

Clinical Characteristics before and after Propensity Score Matching when extubation failure if defined as 72 hours

acc-004250-Supplementary-Table-2.pdf

Figure 1.

Study design. The study phases are presented in accordance with the Consolidated Standards of Reporting Trials (CONSORT) guidelines. EHR: Electronic Health Registry.

Figure 2.

Clinical characteristics before and after propensity score matching. Thirty-day survival analysis (from the time of planned extubation) in those who were successfully weaned off invasive mechanical ventilation (blue) and those with failed extubation (yellow) in the entire cohort (A) as well as the propensity score matched sample (B).

Figure 3.

Average 30-day survival. The primary outcome of average 30-day survival from the initial planned extubation for all patients extubated in a given calendar month is plotted against the proportion of failed planned extubations for the same group. In other words, each dot represents the aggregate extubation failure and average 30-day survival for all patients extubated in the same month. A yellow line delineates the second order regression line fitted for this data, with a bold point indicating the maximum.

Figure 4.

Average ventilation-free days. The secondary outcome of ventilation free days, defined as the number of days each patient was free from any invasive ventilation (including tracheostomy of any sort but excluding noninvasive positive pressure or high flow ventilatory support) for each month is plotted against the average extubation failure rate for this month.

Table 1.

Study patients characteristics

Characteristics	Successful weaning (n=516)	Extubation failure (n=262)	P-value
Age (yr)	56 ±18	63±15	<0.001^a)
Female sex	181 (35.1)	96 (37.6)	0.67^b)
Main indication for invasive ventilation
Acute decompensated heart failure	80 (15.5)	41 (15.6)	0.96^b)
Acute exacerbation of chronic obstructive pulmonary disease	90 (17.4)	35 (13.4)	0.14^b)
Acute kidney injury	34 (6.6)	12 (4.6)	0.26^b)
Pneumonia	152 (29.5)	76 (29.0)	0.90^b)
Other sepsis	105 (20.3)	50 (19.1)	0.68^b)
Trauma	14 (2.7)	2 (0.8)	0.07^c)
Other surgery	35 (6.8)	19 (7.3)	0.81^b)
Stroke	43 (8.3)	15 (5.7)	0.19^b)
Other specific neurological or neuromuscular disorder	27 (5.2)	8 (3.1)	0.17^b)
Other loss of consciousness (excluding cardiac arrest)	103 (20.0)	42 (16.0)	0.18^b)
Cardio-pulmonary resuscitation	10 (1.9)	2 (0.8)	0.36^c)
Medical background
Ischemic heart disease	70 (13.6)	48 (18.3)	0.08^b)
Heart failure	75 (14.5)	45 (17.2)	0.19^b)
Smoking	149 (28.9)	73 (27.9)	0.47^b)
Acute leukemia	6 (0.9)	7 (2.7)	0.05^c)
Other active malignancy	65 (12.6)	48 (18.3)	0.55^b)
Cirrhosis	7 (1.4)	8 (3.1)	0.10^c)
Chronic obstructive pulmonary disease	101(19.6)	42 (16.0)	0.40^b)
Asthma	34 (6.6)	12 (4.6)	0.26^b)
Chronic renal replacement therapy	11 (2.1)	5 (1.9)	0.84^c)

Values are presented as mean±standard deviation or number (%). The baseline characteristics of patients included in the final analysis are presented.

^a)Wilcoxon rank sum test;

^b)Pearson’s chi-squared test;

^c)Fisher’s exact test.

Table 2.

Clinical characteristics before and after propensity score matching

Characteristic	Unmatched			Propensity score matched
Characteristic	Successful weaning (n=516)	Extubation failure (n=262)	P-value^a)	Matched successful weaning (n=262)	Extubation failure (n=262)	P-value^a)
Mean arterial pressure (mm Hg)	91 (76–109)	85 (69–106)	0.004	91 (75–108)	85 (69–107)	0.06
Heart rate (/min)	78 (67–90)	93 (79–110)	<0.001	89 (77–99)	94 (79–110)	0.17
Respiratory rate (/min)	17 (8–28)	17 (15–22)	0.06	17 (8–28)	17 (15–21)	0.19
pH	7.39 (7.36–7.43)	7.39 (7.35–7.42)	0.63	7.40 (7.36–7.43)	7.39 (7.35–7.43)	0.23
Serum sodium (mEq/L)	140.0 (137.0–143.0)	141.0 (136.0–145.0)	0.11	140.0 (137.0–143.0)	141.0 (136.0–145.0)	0.34
Serum potassium (mEq/L)	3.80 (3.70–3.90)	3.80 (3.40–4.18)	0.28	3.80 (3.70–3.90)	3.70 (3.40–4.10)	0.14
Serum creatinine (mg/dl)	1.04 (0.23–2.67)	0.86 (0.64–1.67)	0.04	0.68 (0.23–2.09)	0.80 (0.61–1.56)	0.23
Hemoglobin (mg/dl)	9.90 (8.40–11.60)	9.50 (8.43–11.40)	0.16	9.90 (8.50–11.60)	9.90 (8.68–11.53)	0.76
White blood cells count (×10⁶/µl)	12 (1–28)	13 (9–17)	0.10	9 (1–24)	13 (9–16)	0.13
Platelets (×10⁶/µl)	221 (151–291)	212 (141–302)	0.71	241 (161–303)	212 (143–301)	0.14
Total bilirubin (mg/dl)	0.80 (0.05–1.63)	0.61 (0.36–1.11)	0.44	0.60 (0.05–1.60)	0.58 (0.36–1.03)	0.14
SOFA score at intubation	7.00 (6.00–8.00)	7.00 (6.00–9.00)	0.001	7.00 (6.00–8.00)	7.00 (6.00–9.00)	0.18
Duration of invasive mechanical ventilation (hr)	74 (34–155)	82 (40-142)	0.52	74 (33–155)	82 (40-142)	0.61
SOFA score at extubation	6.00 (4.00–7.00)	7.00 (6.00–8.00)	<0.001	6.50 (5.00–8.00)	7.00 (6.00–8.00)	0.11

Values are presented as median (interquartile range). The clinical characteristics of patients with failed planned extubation are compared with those who underwent successful weaning on the left hand side of this table. On the right, the same characteristics are compared following a 1:1 matching of the propensity for extubation failure. All characteristics except SOFA score are those recorded at the time of extubation.

SOFA: Sequential Organ Failure Assessment.

^a)Wilcoxon rank sum test.

REFERENCES

1. Quintard H, l'Her E, Pottecher J, Adnet F, Constantin JM, De Jong A, et al. Experts' guidelines of intubation and extubation of the ICU patient of French Society of Anaesthesia and Intensive 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 2019;9:13.Article PubMed PMC
2. Pham T, Heunks L, Bellani G, Madotto F, Aragao I, Beduneau G, et al. Weaning from mechanical ventilation in intensive care units across 50 countries (WEAN SAFE): a multicentre, prospective, observational cohort study. Lancet Respir Med 2023;11:465-76.Article PubMed
3. Krinsley JS, Reddy PK, Iqbal A. What is the optimal rate of failed extubation? Crit Care 2012;16:111.Article PubMed PMC
4. Igarashi Y, Ogawa K, Nishimura K, Osawa S, Ohwada H, Yokobori S, et al. Machine learning for predicting successful extubation in patients receiving mechanical ventilation. Front Med (Lausanne) 2022;9:961252.Article PubMed PMC
5. Chen X, Li J, Liu G, Chen X, Huang S, Li H, et al. Identification of distinct clinical phenotypes of heterogeneous mechanically ventilated ICU patients using cluster analysis. J Clin Med 2023;12:1499.Article PubMed PMC
6. Tanaka A, Kabata D, Hirao O, Kosaka J, Furushima N, Maki Y, et al. Prediction model of extubation outcomes in critically ill patients: a multicenter prospective cohort study. J Clin Med 2022;11:2520.Article PubMed PMC
7. Thille AW, Muller G, Gacouin A, Coudroy R, Decavèle M, Sonneville R, et al. Effect of postextubation high-flow nasal oxygen with noninvasive ventilation vs high-flow nasal oxygen alone on reintubation among patients at high risk of extubation failure: a randomized clinical trial. JAMA 2019;322:1465-75.Article PubMed PMC
8. Damuth E, Mitchell JA, Bartock JL, Roberts BW, Trzeciak S. Long-term survival of critically ill patients treated with prolonged mechanical ventilation: a systematic review and meta-analysis. Lancet Respir Med 2015;3:544-53.Article PubMed
9. Zilberberg MD, Nathanson BH, Ways J, Shorr AF. Characteristics, hospital course, and outcomes of patients requiring prolonged acute versus short-term mechanical ventilation in the united states, 2014-2018. Crit Care Med 2020;48:1587-94.Article PubMed
10. Trivedi V, Chaudhuri D, Jinah R, Piticaru J, Agarwal A, Liu K, et al. The usefulness of the rapid shallow breathing index in predicting successful extubation: a systematic review and meta-analysis. Chest 2022;161:97-111.Article PubMed
11. Baptistella AR, Mantelli LM, Matte L, Carvalho ME, Fortunatti JA, Costa IZ, et al. Prediction of extubation outcome in mechanically ventilated patients: development and validation of the Extubation Predictive Score (ExPreS). PLoS One 2021;16:e0248868. Article PubMed PMC
12. Kapnadak SG, Herndon SE, Burns SM, Shim YM, Enfield K, Brown C, et al. Clinical outcomes associated with high, intermediate, and low rates of failed extubation in an intensive care unit. J Crit Care 2015;30:449-54.Article PubMed
13. MacIntyre NR, Cook DJ, Ely EW, Epstein SK, Fink JB, Heffner JE, et al. Evidence-based guidelines for weaning and discontinuing ventilatory support: a collective task force facilitated by the American College of Chest Physicians; the American Association for Respiratory Care; and the American College of Critical Care Medicine. Chest 2001;120:375S-95S.Article PubMed
14. Torrini F, Gendreau S, Morel J, Carteaux G, Thille AW, Antonelli M, et al. Prediction of extubation outcome in critically ill patients: a systematic review and meta-analysis. Crit Care 2021;25:391.Article PubMed PMC PDF
15. Béduneau G, Pham T, Schortgen F, Piquilloud L, Zogheib E, Jonas M, et al. Epidemiology of weaning outcome according to a new definition. The WIND study. Am J Respir Crit Care Med 2017;195:772-83.Article PubMed
16. Esteban A, Anzueto A, Frutos F, Alía I, Brochard L, Stewart TE, et al. Characteristics and outcomes in adult patients receiving mechanical ventilation: a 28-day international study. JAMA 2002;287:345-55.Article PubMed
17. Russotto V, Myatra SN, Laffey JG, Tassistro E, Antolini L, Bauer P, et al. INTUBE Study Investigators. Intubation practices and adverse peri-intubation events in critically ill patients from 29 countries. JAMA 2021;325:1164-1172.Article PubMed PMC
18. Ding C, Zhang Y, Yang Z, Wang J, Jin A, Wang W, et al. Incidence, temporal trend and factors associated with ventilator-associated pneumonia in mainland China: a systematic review and meta-analysis. BMC Infect Dis 2017;17:468.Article PubMed PMC PDF
19. Gunalan A, Sastry AS, Ramanathan V, Sistla S. Early- vs late-onset ventilator-associated pneumonia in critically ill adults: comparison of risk factors, outcome, and microbial profile. Indian J Crit Care Med 2023;27:411-5.Article PubMed PMC
20. Epstein SK, Ciubotaru RL, Wong JB. Effect of failed extubation on the outcome of mechanical ventilation. Chest 1997;112:186-92.Article PubMed
21. Seymour CW, Martinez A, Christie JD, Fuchs BD. The outcome of extubation failure in a community hospital intensive care unit: a cohort study. Crit Care 2004;8:R322-7.Article PubMed PMC PDF
22. Gao F, Yang LH, He HR, Ma XC, Lu J, Zhai YJ, et al. The effect of reintubation on ventilator-associated pneumonia and mortality among mechanically ventilated patients with intubation: a systematic review and meta-analysis. Heart Lung 2016;45:363-71.Article PubMed
23. Hernández Martínez G, Rodriguez P, Soto J, Caritg O, Castellví-Font A, Mariblanca B, et al. Effect of aggressive vs conservative screening and confirmatory test on time to extubation among patients at low or intermediate risk: a randomized clinical trial. Intensive Care Med 2024;50:258-67.Article PubMed PDF
24. Boles JM, Bion J, Connors A, Herridge M, Marsh B, Melot C, et al. Weaning from mechanical ventilation. Eur Respir J 2007;29:1033-56.Article PubMed
25. Tanaka A, Shimomura Y, Uchiyama A, Tokuhira N, Kitamura T, Iwata H, et al. Time definition of reintubation most relevant to patient outcomes in critically ill patients: a multicenter cohort study. Crit Care 2023;27:378.Article PubMed PMC PDF
26. Miltiades AN, Gershengorn HB, Hua M, Kramer AA, Li G, Wunsch H, et al. Cumulative probability and time to reintubation in U.S. ICUs. Crit Care Med 2017;45:835-42.Article PubMed PMC
27. Thille AW, Richard JC, Brochard L. The decision to extubate in the intensive care unit. Am J Respir Crit Care Med 2013;187:1294-302.Article PubMed
28. Ferreira FL, Bota DP, Bross A, Mélot C, Vincent JL. Serial evaluation of the SOFA score to predict outcome in critically ill patients. JAMA 2001;286:1754-8.Article PubMed

Figure & Data

References

Citations

Citations to this article as recorded by

PubReader
ePub Link

Cite

CITE: export Copy Format; Close

Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:

RIS — For EndNote, ProCite, RefWorks, and most other reference management software
BibTeX — For JabRef, BibDesk, and other BibTeX-specific software

Include:

Citation for the content below
Citation and abstract for the content below

Association of extubation failure rate with patient outcomes

Acute Crit Care. 2026;41(2):295-303. Published online April 17, 2026

DOI: https://doi.org/10.4266/acc.004250

XML Download

Figure

Related articles

Duration of antibiotic therapy: with or without biomarkers?

Association of extubation failure rate with patient outcomes

Figure 1. Study design. The study phases are presented in accordance with the Consolidated Standards of Reporting Trials (CONSORT) guidelines. EHR: Electronic Health Registry.

Figure 2. Clinical characteristics before and after propensity score matching. Thirty-day survival analysis (from the time of planned extubation) in those who were successfully weaned off invasive mechanical ventilation (blue) and those with failed extubation (yellow) in the entire cohort (A) as well as the propensity score matched sample (B).

Figure 3. Average 30-day survival. The primary outcome of average 30-day survival from the initial planned extubation for all patients extubated in a given calendar month is plotted against the proportion of failed planned extubations for the same group. In other words, each dot represents the aggregate extubation failure and average 30-day survival for all patients extubated in the same month. A yellow line delineates the second order regression line fitted for this data, with a bold point indicating the maximum.

Figure 4. Average ventilation-free days. The secondary outcome of ventilation free days, defined as the number of days each patient was free from any invasive ventilation (including tracheostomy of any sort but excluding noninvasive positive pressure or high flow ventilatory support) for each month is plotted against the average extubation failure rate for this month.

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Association of extubation failure rate with patient outcomes

Characteristics	Successful weaning (n=516)	Extubation failure (n=262)	P-value
Age (yr)	56 ±18	63±15	<0.001^a)
Female sex	181 (35.1)	96 (37.6)	0.67^b)
Main indication for invasive ventilation
Acute decompensated heart failure	80 (15.5)	41 (15.6)	0.96^b)
Acute exacerbation of chronic obstructive pulmonary disease	90 (17.4)	35 (13.4)	0.14^b)
Acute kidney injury	34 (6.6)	12 (4.6)	0.26^b)
Pneumonia	152 (29.5)	76 (29.0)	0.90^b)
Other sepsis	105 (20.3)	50 (19.1)	0.68^b)
Trauma	14 (2.7)	2 (0.8)	0.07^c)
Other surgery	35 (6.8)	19 (7.3)	0.81^b)
Stroke	43 (8.3)	15 (5.7)	0.19^b)
Other specific neurological or neuromuscular disorder	27 (5.2)	8 (3.1)	0.17^b)
Other loss of consciousness (excluding cardiac arrest)	103 (20.0)	42 (16.0)	0.18^b)
Cardio-pulmonary resuscitation	10 (1.9)	2 (0.8)	0.36^c)
Medical background
Ischemic heart disease	70 (13.6)	48 (18.3)	0.08^b)
Heart failure	75 (14.5)	45 (17.2)	0.19^b)
Smoking	149 (28.9)	73 (27.9)	0.47^b)
Acute leukemia	6 (0.9)	7 (2.7)	0.05^c)
Other active malignancy	65 (12.6)	48 (18.3)	0.55^b)
Cirrhosis	7 (1.4)	8 (3.1)	0.10^c)
Chronic obstructive pulmonary disease	101(19.6)	42 (16.0)	0.40^b)
Asthma	34 (6.6)	12 (4.6)	0.26^b)
Chronic renal replacement therapy	11 (2.1)	5 (1.9)	0.84^c)

Characteristic	Unmatched			Propensity score matched
Characteristic	Successful weaning (n=516)	Extubation failure (n=262)	P-value^a)	Matched successful weaning (n=262)	Extubation failure (n=262)	P-value^a)
Mean arterial pressure (mm Hg)	91 (76–109)	85 (69–106)	0.004	91 (75–108)	85 (69–107)	0.06
Heart rate (/min)	78 (67–90)	93 (79–110)	<0.001	89 (77–99)	94 (79–110)	0.17
Respiratory rate (/min)	17 (8–28)	17 (15–22)	0.06	17 (8–28)	17 (15–21)	0.19
pH	7.39 (7.36–7.43)	7.39 (7.35–7.42)	0.63	7.40 (7.36–7.43)	7.39 (7.35–7.43)	0.23
Serum sodium (mEq/L)	140.0 (137.0–143.0)	141.0 (136.0–145.0)	0.11	140.0 (137.0–143.0)	141.0 (136.0–145.0)	0.34
Serum potassium (mEq/L)	3.80 (3.70–3.90)	3.80 (3.40–4.18)	0.28	3.80 (3.70–3.90)	3.70 (3.40–4.10)	0.14
Serum creatinine (mg/dl)	1.04 (0.23–2.67)	0.86 (0.64–1.67)	0.04	0.68 (0.23–2.09)	0.80 (0.61–1.56)	0.23
Hemoglobin (mg/dl)	9.90 (8.40–11.60)	9.50 (8.43–11.40)	0.16	9.90 (8.50–11.60)	9.90 (8.68–11.53)	0.76
White blood cells count (×10⁶/µl)	12 (1–28)	13 (9–17)	0.10	9 (1–24)	13 (9–16)	0.13
Platelets (×10⁶/µl)	221 (151–291)	212 (141–302)	0.71	241 (161–303)	212 (143–301)	0.14
Total bilirubin (mg/dl)	0.80 (0.05–1.63)	0.61 (0.36–1.11)	0.44	0.60 (0.05–1.60)	0.58 (0.36–1.03)	0.14
SOFA score at intubation	7.00 (6.00–8.00)	7.00 (6.00–9.00)	0.001	7.00 (6.00–8.00)	7.00 (6.00–9.00)	0.18
Duration of invasive mechanical ventilation (hr)	74 (34–155)	82 (40-142)	0.52	74 (33–155)	82 (40-142)	0.61
SOFA score at extubation	6.00 (4.00–7.00)	7.00 (6.00–8.00)	<0.001	6.50 (5.00–8.00)	7.00 (6.00–8.00)	0.11

Table 1. Study patients characteristics

Values are presented as mean±standard deviation or number (%). The baseline characteristics of patients included in the final analysis are presented.

Wilcoxon rank sum test;

Pearson’s chi-squared test;

Fisher’s exact test.

Table 2. Clinical characteristics before and after propensity score matching

SOFA: Sequential Organ Failure Assessment.

Wilcoxon rank sum test.