Postextubation management in head injury patients at high risk of reintubation in the trauma intensive care unit: a retrospective study

Abstract

Purpose

Careful clinical assessment is essential for extubation in patients with acute respiratory failure, and decisions must weigh the potential benefits and risks related to the timing of extubation. This study aimed to characterize the population at high risk for extubation failure and to assess the interventions employed to prevent reintubation in head injury patients admitted to the trauma intensive care unit.

Methods

We retrospectively obtained data from the trauma registry database of traumatic brain injury patients who were intubated and admitted to a trauma intensive care unit in Qatar from January 2013 to December 2015. Regression analysis was performed to examine the association between successful extubation and in-hospital complications.

Results

In total, 297 head injury patients (279 male and 26 female patients) at high risk of reintubation were included in the analysis. The prevalence of agitation and pneumonia in postextubated head injury patients were 10.8% and 4.7%, respectively. After adjusting for confounding variables, successful extubation with the use of a nasal cannula and continuous positive airway pressure was significantly associated with reduced agitation (P=0.001) and a lower incidence of in-hospital complications, including pneumonia (P=0.001) and sepsis (P=0.02).

Conclusions

This study suggests an association between successful extubation, reduced agitation, and a decreased incidence of in-hospital complications such as pneumonia and sepsis. Particular attention and further research should focus on optimizing extubation management for head injury patients.

INTRODUCTION

Background

A careful clinical assessment is essential for extubation among patients with acute respiratory failure, and decisions should be based on a thorough consideration of the benefits and risks associated with the timing of extubation [1]. These patients require close monitoring to ensure significant resolution of the underlying condition causing respiratory failure, along with satisfactory oxygenation and appropriate weaning parameters [2,3]. Only a few studies have developed predictive indices for successful extubation in patients with impaired mental status. Nonetheless, clinical experience shows that many patients with severely abnormal mental status can be safely managed without an artificial airway. In the absence of clear guidelines, clinical practices often vary significantly [4].

Some patients may also have significant preexisting respiratory impairment, complicating baseline assessments. Reintubation in these patients leads to increased morbidity and mortality, with rates exceeding 40% in some reports [5,6]. Thus, identifying patients at high risk for extubation failure and implementing effective postextubation management strategies are critical to improving clinical outcomes.

Postextubation management strategies focus on both preventing reintubation and maintaining patient comfort. Conventional oxygenation via face masks has long been standard, but alternative approaches such as noninvasive ventilation (NIV) and high-flow nasal cannula (HFNC) oxygenation are now used in many clinical settings, particularly for acute hypoxemic respiratory failure [6,7]. Recent research has highlighted the benefits of HFNC in lowering reintubation rates compared to conventional oxygen therapy. For example, in a randomized controlled trial, Hernández et al. [7] compared NIV with HFNC after extubation in high-risk patients recovering from acute respiratory failure and found a significantly lower reintubation rate within 72 hours in the HFNC group (4.9%) versus conventional oxygen therapy (12.2%). Moreover, the time to reintubation was longer in the HFNC group (19 hours) than in the conventional therapy group (15 hours).

Similarly, Maggiore et al. [8] evaluated HFNC versus Venturi mask oxygen in 105 postextubation patients and observed that HFNC was associated with significantly lower reintubation rates (4% vs. 21%), higher PaO₂ to fraction of inspired oxygen (FiO₂) ratios, and improved patient comfort. These findings suggest that HFNC reduces reintubation and improves comfort in low-risk patients. Two recent meta-analyses provide further insights into NIV in postextubation settings. Glossop et al. [9] analyzed four randomized controlled trials with 479 postextubation patients and found an odds ratio for reintubation of 0.72 (95% confidence interval, 0.51–1.02; P=0.28) for those using NIV. Lin et al. [6] examined eight trials (1,080 patients) randomized to standard medical therapy or NIV following extubation, finding no reduction in reintubation rates with NIV. These meta-analyses indicate that information on the effectiveness of NIV immediately postextubation remains limited [9–11]. The impact of extubation failure on outcomes in patients with traumatic brain injury is also not well understood and may be underestimated. Given the high prevalence of traumatic brain injury and the potential adverse consequences of extubation failure, there is an urgent need to better characterize high-risk populations and evaluate postextubation strategies in this vulnerable group.

Objectives

This study aims to address this knowledge gap by characterizing patients at high risk of extubation failure and evaluating the maneuvers used to avoid reintubation in traumatic brain injury patients admitted to the trauma intensive care unit (TICU).

METHODS

Ethics statement

The study protocol was reviewed and approved by the Institutional Review Board of Hamad Medical Corporation (No. MRC 01-20-122). The requirement for informed consent was waived because all data were collected anonymously and de-identified without patient contact. All study procedures were conducted in accordance with the principles of the Declaration of Helsinki.

Study design and setting

This retrospective cohort study included adult patients with traumatic brain injury who were intubated and admitted to the TICU at the level I trauma center of Hamad General Hospital (Doha, Qatar) from January 2013 to December 2015. Adult patients with head injuries who were intubated and hospitalized were eligible for inclusion.

Data collection

Data were obtained from the trauma registry database. Patients with preexisting medical conditions affecting respiratory function, such as asthma or chronic obstructive pulmonary disease, were excluded to minimize potential confounding effects. Variables collected included demographic characteristics (age, sex, nationality), mechanism of injury (blunt or penetrating), associated injuries by anatomical region (chest, spine, pelvis, abdomen, extremities, etc.), reintubation, agitation, self-extubation, use of analgesics and sedatives, FiO₂, use of continuous positive airway pressure (CPAP) or HFNC, Ramsay sedation scale, Injury Severity Score, Glasgow Coma Scale (GCS) score at TICU admission, head Abbreviated Injury Score, ventilatory days, and hospital and ICU length of stay. Postextubation airway compromise (respiratory failure) was defined as the presence and persistence of any of the following: respiratory acidosis (pH <7.35), oxygen saturation as measured by pulse oximetry (SpO₂) <90% or PaO₂ (35–45 mmHg) to FiO₂ ratio >0.4, respiratory rate >35 breaths per minute, decreased level of consciousness, agitation, or clinical signs of respiratory muscle fatigue and increased work of breathing. Reintubation was performed when one or more of the following occurred within 48 hours after extubation: clinical signs of increased respiratory effort, upper airway obstruction, respiratory acidosis, hypoxemia (PaO₂ <60 mmHg), decreased consciousness (GCS) with inability to protect the airway, tachycardia, desaturation, or tachypnea. The primary outcome was postextubation respiratory failure (reintubation) in the TICU.

Our institution utilizes a standardized mechanical ventilator weaning protocol. Weaning begins with a gradual reduction in the mandatory ventilation rate to encourage spontaneous breathing, followed by titration of pressure support to strengthen respiratory effort while maintaining stability. Patients then undergo T-piece trials to assess their ability to breathe spontaneously without ventilatory support. Finally, they are transitioned to spontaneous breathing with minimal pressure support and positive end-expiratory pressure to confirm readiness for extubation. This weaning process is uniformly applied, with adjustments only as clinically indicated.

For pain, agitation, and delirium syndrome, our management protocol uses a structured pharmacological approach, primarily including β-blockers, benzodiazepines, and baclofen, with the selective addition of clonidine when clinically appropriate.

Statistical analysis

Data are presented as proportions (%), mean±standard deviation for normally distributed continuous variables, or median with interquartile range or range for non-normally distributed variables. Baseline demographics, clinical presentation, management, and outcomes were compared between the reintubation and successful extubation groups. The Student t-test was used for normally distributed continuous variables, and the Mann-Whitney U-test for non-normally distributed variables (e.g., ICU length of stay, ventilatory days). Categorical variables were analyzed using the Pearson chi-square test or the Fisher exact test where expected cell counts were <5.

Variables that were significant in univariate analysis or considered clinically relevant were entered into a multivariate logistic regression model to identify independent risk factors for reintubation. A two-tailed P-value of <0.05 was considered statistically significant. Data analysis was performed using IBM SPSS ver. 22 (IBM Corp).

RESULTS

Patient demographics and injury characteristics

The study included 297 head injury patients at high risk of reintubation in the TICU. Fig. 1 presents the study flowchart, detailing participant enrollment and associated clinical outcomes after extubation. The cohort was predominantly male (271 patients, 91.2%), with an average age of 33.2±12.6 years. Traffic accidents were the most common mechanism of injury (198 patients, 66.7%), followed by falls (72 patients, 24.2%). The leading types of head injuries included brain contusions (188 patients, 63.3%), skull fractures (179 patients, 60.3%), and subdural hematomas (111 patients, 37.4%). In total, 18 patients (6.1%) required reintubation after extubation (Table 1).

Fig. 1.

Flowchart of participants enrolled in the study and their clinical outcomes after extubation. TICU, trauma intensive care unit.

Table 1.

Postextubation management of head injury patients at high-risk of reintubation in the TICU (n=297)

Variable	Value
Age (yr)	33.2±12.6
Sex
Male	271 (91.2)
Female	26 (8.8)
Mechanism of injury
Traffic accident	198 (66.7)
Fall	72 (24.2)
Fall of heavy objects	6 (2.0)
All-terrain vehicle	4 (1.3)
Assault	7 (2.4)
Other	10 (3.4)
Types of head injury
Skull fracture	179 (60.3)
Brain contusion	188 (63.3)
Subdural hematoma	111 (37.4)
Epidural hematoma	63 (21.2)
Subarachnoid hemorrhage	109 (36.7)
Intraventricular hemorrhage	39 (13.1)
Extra-axial hematoma	20 (6.7)
Pneumocephalus	43 (14.5)
Brain edema	109 (36.7)
Diffuse axonal injury	62 (20.9)
GCS score upon TICU admission	10 (3–15)
Head AIS score	4.0±1.0
Injury Severity Score	28.4±10.4
Tracheostomy	83 (27.9)
Analgesia	253 (85.2)
Fentanyl	215 (85.0)
Remifentanil	38 (15.0)
Sedation	244 (82.2)
Propofol	175 (71.7)
Midazolam	69 (28.3)
Haloperidol	83 (27.9)
Quetiapine	61 (20.5)
Dexmedetomidine	30 (10.1)
Tramadol	16 (5.4)
FiO2 (%) (n=247)a)	48.6±19.8
High-flow nasal cannula	99 (33.3)
Face mask	97 (32.7)
Continuous positive airway pressure	51 (17.2)
Reintubation	18 (6.1)
Neurosurgical interventionb)	71 (23.9)
TICU LOS (day)	19 (5–163)
Mechanical ventilation use (day)	12 (5–47)
Hospital LOS (day)	31 (5-361)
In-hospital complication
Agitation	32 (10.8)
Pneumonia	14 (4.7)
Acute respiratory distress syndrome	22 (7.4)
Sepsis	67 (22.6)
Mortality	55 (18.5)

Values are presented as mean±standard deviation, number (%), or median (range).

TICU, trauma intensive care unit; GCS, Glasgow Coma Score; AIS, Abbreviated Injury Scale; FiO₂, fraction of inspired oxygen; LOS, length of stay.

^a)Data on FiO₂ were not available for 50 patients.

^b)Craniotomy/craniectomy.

Postextubation management

Analgesia and sedation were frequently administered to extubated patients, with 253 patients (85.2%) receiving analgesia and 244 patients (82.2%) requiring sedation for discomfort and agitation. Fentanyl was the primary analgesic used (215 patients, 85.0%), followed by remifentanil (38 patients, 15.0%). Among sedatives, propofol was most common (175 patients, 71.7%), with midazolam (69 patients, 28.3%), haloperidol (83 patients, 27.9%), and quetiapine (61 patients, 20.5%) also used, reflecting a preference for agents compatible with ongoing neurological assessment in head injury patients (Table 1).

The mean FiO₂ was 48.6%±19.8%. Oxygen was administered via several methods, with a HFNC used most frequently (99 patients, 33.3%), closely followed by a face mask (97 patients, 32.7%). CPAP was used in 51 patients (17.2%), particularly where higher respiratory support was needed. Neurosurgical intervention was performed in 71 patients (23.9%). Mechanical ventilation was required for a median of 12 days, and 27.9% of patients underwent tracheostomy. Complications after extubation included agitation in 32 patients (10.8%), pneumonia in 14 (4.7%), and acute respiratory distress syndrome in 22 (7.4%). Sepsis was the most prevalent hospital complication, affecting 67 patients (22.6%). The overall mortality rate in the cohort was 18.5% (Table 1).

Comparison between successful extubation and reintubation groups

Table 2 summarizes factors associated with successful extubation and reintubation. Of the 297 patients, 279 (93.9%) achieved successful extubation, while 18 (6.1%) required reintubation. There were no significant differences in age, sex, or Injury Severity Score between groups.

Table 2.

Reintubation in the TICU (n=297)

Variable	Successful extubation (n=279)	Reintubation (n=18)	P-value
Age (yr)	32.1±13.5	35.5±10.6	0.31
Male sex	253 (90.7)	18 (100)	0.17
Type of head injury
Skull fracture	168 (60.2)	11 (61.1)	0.94
Brain contusion	176 (63.1)	12 (66.7)	0.76
Subdural hematoma	101 (36.2)	10 (55.6)	0.10
Epidural hematoma	58 (20.8)	5 (27.8)	0.48
Subarachnoid hemorrhage	100 (35.8)	9 (50.0)	0.22
Intraventricular hemorrhage	38 (13.6)	1 (5.6)	0.32
Brain edema	105 (37.6)	4 (22.2)	0.18
Diffuse axonal injury	57 (20.4)	5 (27.8)	0.45
GCS score upon TICU admission	10 (3–15)	10 (6–11)	0.13
Head AIS score	4.0±1.0	4.2±1.0	0.40
Injury Severity Score	28.4±10.4	29.7±10.5	0.62
Tracheostomy	78 (28.0)	5 (27.8)	0.98
Neurosurgical interventiona)	65 (23.3)	6 (33.3)	0.33
Analgesia	235 (84.2)	18 (100)	0.06
Fentanyl	197 (83.8)	18 (100)	0.06
Remifentanil	38 (16.2)	0	0.06
Sedation	226 (81.0)	18 (100)	0.04*
Propofol	159 (70.4)	16 (88.9)	0.09
Midazolam	67 (29.6)	2 (11.1)	0.09
Haloperidol	75 (26.9)	8 (44.4)	0.10
Quetiapine	56 (20.1)	5 (27.8)	0.43
Dexmedetomidine	26 (9.3)	4 (22.2)	0.17
Tramadol	9 (3.2)	7 (38.9)	0.001*
FiO2 (%) (n=247)b)	46.5±18.3	75.0±18.3	0.001*
High-flow nasal cannula	98 (35.1)	1 (5.6)	0.02*
Face mask	94 (33.7)	3 (16.7)	0.13
Continuous positive airway pressure	37 (13.3)	14 (77.8)	0.001*
TICU LOS (day)	19 (5–163)	21 (13–58)	0.18
Mechanical ventilation use (day)	12 (5–47)	14 (5–26)	0.19
Hospital LOS (day)	31 (5–361)	42 (15–91)	0.11
In-hospital complication
Agitation	22 (7.9)	10 (55.6)	0.001*
Pneumonia	2 (0.7)	12 (66.7)	0.001*
Acute respiratory distress syndrome	21 (7.5)	1 (5.6)	0.87
Sepsis	59 (21.1)	8 (44.4)	0.02*
Mortality	54 (19.4)	1 (5.6)	0.25

Values are presented as mean±standard deviation, number (%), or median (range).

TICU, trauma intensive care unit; GCS, Glasgow Coma Score; AIS, Abbreviated Injury Scale; FiO₂, fraction of inspired oxygen; LOS, length of stay.

^a)Craniotomy/craniectomy.

^b)Data on FiO₂ were not available for 50 patients.

^*P<0.05.

Patients who required reintubation had significantly higher rates of agitation (55.6% vs. 7.9%, P=0.001), pneumonia (66.7% vs. 0.7%, P=0.001), and sepsis (44.4% vs. 21.1%, P=0.02) compared to those with successful extubation. Sedation was used in all reintubated patients (100%) versus 226 (81.0%) in the successful extubation group (P=0.04). Tramadol use was also significantly higher in the reintubation group (38.9% vs. 3.2%, P=0.001). Notably, patients who required reintubation had significantly higher FiO₂ levels (75.0%±18.3% vs. 46.5%±18.3%, P=0.001), reflecting more severe respiratory compromise after extubation. CPAP was also used more frequently among reintubated patients (77.8% vs. 13.3%, P=0.001). Although neurosurgical intervention rates were higher in reintubated patients, this difference was not statistically significant (33.3% vs. 23.3%, P=0.33).

TICU and hospital stay outcomes

The median ICU length of stay was 19 days (range, 5–163 days), and median hospital length of stay was 31 days (range, 5–361 days). Patients requiring reintubation had a longer median hospital stay (42 days), though this was not statistically significant (P=0.11). The mortality rate for the overall cohort was 18.5%, with no significant difference between the successful extubation and reintubation groups (19.4% vs. 5.6%, P=0.25).

DISCUSSION

This study examined postextubation management and outcomes in head injury patients at high risk for reintubation in a TICU. The results demonstrate that successful extubation is significantly associated with lower rates of agitation and fewer in-hospital complications such as pneumonia and sepsis. These findings support earlier research highlighting the importance of tailored postextubation strategies to minimize complications and improve outcomes. The reintubation rate observed in this study was 6.1%, which is lower than rates reported in similar populations, where reintubation frequently exceeds 10% [1,2]. This discrepancy may reflect improvements in clinical protocols and the use of evidence-based interventions such as HFNC oxygen therapy and continuous CPAP, though it could also be influenced by factors such as patient demographics, initial injury severity, and baseline comorbidities.

For example, Hernández et al. [7] found that HFNC was superior to NIV in reducing reintubation rates (4.9% vs. 12.2%) and related respiratory failure. The significant association in our study between CPAP usage and successful extubation aligns with these findings, emphasizing CPAP’s role in optimizing oxygenation and decreasing respiratory effort. Agitation emerged as a key predictor of reintubation, with those requiring reintubation experiencing substantially higher agitation rates (55.6% vs. 7.9%). This underscores the necessity for effective sedation management in head injury patients. Previous studies have also identified agitation as a risk factor for extubation failure, highlighting the need to balance adequate sedation with preservation of neurological function [3,4]. Our results indicate that sedation protocols utilizing agents like propofol and dexmedetomidine may improve outcomes, as their use was more common among those who were successfully extubated.

Pneumonia was a particularly prominent complication among re-intubated patients (66.7% vs. 0.7%, P=0.001), echoing prior research that identifies pneumonia as a major consequence of extubation failure in trauma and neurological injury populations [5,6]. Enhanced airway management techniques, such as HFNC and CPAP, may mitigate the risk of pneumonia by improving mucociliary clearance and reducing the need for invasive reintubation. These strategies may also improve patient comfort, as seen in Maggiore et al. [8], where HFNC was associated with better oxygenation and greater patient comfort compared to Venturi masks.

The relationship between FiO₂ requirements and extubation outcomes deserves attention. Patients who required reintubation had significantly higher FiO₂ levels (75.0%±18.3% vs. 46.5%±18.3%, P=0.001), signifying more severe respiratory compromise. Although high FiO₂ is sometimes necessary to manage acute hypoxemia, it also reflects ongoing challenges in achieving adequate oxygenation in these patients [9,10].

Regarding hospital outcomes, the median TICU length of stay was slightly higher for the reintubation group, although this difference was not statistically significant. Mortality rates were similar between groups, suggesting that prompt and effective postextubation interventions may mitigate the long-term impact of extubation failure. These results emphasize the importance of continuous monitoring and timely management of high-risk patients to prevent adverse outcomes.

The findings are consistent with the expanding evidence base supporting individualized extubation strategies for patients with traumatic brain injuries. Future research should focus on improving risk stratification tools and developing standardized protocols for postextubation care. Additionally, prospective studies are needed to validate these findings and assess novel interventions that may further reduce reintubation rates and complications.

Limitations

This study has some limitations that should be acknowledged. First, its retrospective design may have introduced selection bias, limiting the ability to establish causal relationships. Second, data were collected from a single trauma center, which may limit the generalizability of the findings to other settings. Third, the absence of detailed long-term follow-up data restricts our understanding of patient outcomes beyond hospital discharge. Due to the small sample size in the reintubation group, the study is underpowered, limiting the findings' robustness. Finally, GCS scores were not routinely recorded immediately before extubation, as these cases primarily involve trauma patients whose neurological status may exhibit fluctuations. Despite these limitations, the study has notable strengths, including a well-defined cohort of high-risk head injury patients and comprehensive data on extubation management and associated outcomes. Additionally, the analysis provides valuable practical insights for minimizing postextubation complications, contributing to the evidence base for clinical decision-making.

Conclusions

This study highlights the importance of individualized postextubation management in head injury patients at high risk for reintubation. Successful extubation was significantly associated with reduced agitation and fewer in-hospital complications, including pneumonia and sepsis. The use of CPAP was particularly effective in decreasing reintubation rates, supporting its role as a standard strategy in this population. Additionally, higher FiO₂ requirements and agitation were strong predictors of reintubation, highlighting the need for close monitoring and optimized sedation management. Although the reintubation rate was relatively low, the high complication burden among reintubated patients reinforces the necessity of evidence-based extubation protocols. Future research should aim to refine risk stratification methods and evaluate novel postextubation interventions, such as HFNC and tailored sedation regimens, to further improve outcomes in this vulnerable patient group.