Implications of Extubation Delay in Brain-Injured Patients
Abstract & Commentary
Coplin and colleagues, in this large prospective cohort study from Seattle, were trying to analyze the outcomes of intubated, acutely brain-injured patients in relation to time of extubation. They hypothesized that variation in extubating such patients would affect the incidence of nosocomial pneumonia, ICU length of stay, and hospital charges.
Patients with brain injury, traumatic or nontraumatic, who required mechanical ventilation, were evaluated. Patients who required mechanical ventilation for concurrent non-neurological conditions were excluded. Criteria used to determine readiness for discontinuation of ventilatory support were consistent with published criteria and are shown in the Table. For each patient and daily throughout the study, the Glasgow Coma Scale (GCS) score and a six-part Airway Care Score (ACS), including amount and character of secretions, vigor of gag and spontaneous cough, and required frequency of suctioning, were recorded. Patients were classified to have a delay in extubation if endotracheal intubation was maintained for longer than 48 hours after they first met all of the criteria for readiness for discontinuation of mechanical ventilation shown in the Table.
| Table-Criteria Used to Determine Readiness for Discontinuation of Ventilatory Support (Patients had to meet all criteria) |
|
| Category | Criteria |
| Neurological status | Physical examination not deteriorating ICP < 20 mm Hg when measured CPP at least 60 mm Hg when ICP measured |
| Cardiovascular status | Systolic BP > 90 and < 160 mm Hg Heart rate > 60 and < 125 beats / min No acute dysrhythmia |
| Arterial oxygenation | PaO2/FIO2 at least 200 mm Hg PaO2 at least 80 mm Hg on FIO2 0.50 or less (on PEEP of not more than 5 cm H2O) |
| Spontaneous ventilatory mechanics | MIP > 20 cm H2O RSBI (f /VT) < 105 Ventilation requirement 12 L/min or less Spontaneous ventilation at least 80% of ventilation requirement |
| Absence of specific indication | Surgery requiring general anesthesia for mechanical ventilation not planned within 72 hours No deliberate hyperventilation Cervical-spine status cleared |
| ICP—intracranial pressure; CPP—cerebral perfusion pressure; MIP—maximum inspiratory pressure; RSBI—rapid shallow breathing index; f—respiratory rate; Vt—tidal volume |
|
Of the 242 patients evaluated for the study, 136 were included and 106 were excluded, based on predetermined exclusion criteria. Of the 136 study patients, 99 were extubated without delay as compared to 37 patients who had delay in extubation. In general, the patients with delay in extubation were ventilated for a longer period of time before meeting the readiness criteria, required more airway care, and had lower GCS scores than those without extubation delay; however, 58% of patients meeting readiness criteria after at least five days of mechanical ventilation were extubated without delay, and 22% of the patients meeting the readiness criteria in first three days of starting mechanical ventilation had delay in extubation.
Patients with GCS of 8 or less were more likely to have delay in extubation, although some patients with low GCS scores (£ 4) were extubated without delay. Likewise, patients with higher ACS, meaning poor airway function and secretion clearance, were more likely to have delayed extubation, although some of the patients with a high ACS (³ 10) were extubated without delay.
Reintubation for airway or pulmonary dysfunction was not significantly related to extubation delay or to the presence of coma at the time of extubation. Absence of gag reflex or cough was also not associated with reintubation, and in this study 89% of the patients with absent or weak gag reflex were extubated successfully. Two components of the ACS, the presence of a spontaneous cough and suctioning frequency on the day the weaning criteria were first met, were associated with successful extubation.
Of the 136 patients in the study group, 35 developed pneumonia. Extubation delay was associated with a statistically significant increase in the incidence of pneumonia. Pneumonia did not appear to cause extubation delay, as 11/of 16 patients developing pneumonia prior to meeting readiness criteria were extubated without delay. Of the remaining 19 cases of pneumonia, nine occurred during extubation delay and 10 occurred after extubation. There was a 3.7-fold increase in pneumonia in patients who were comatose on the day they first met criteria for weaning and who also had a delay in extubation. Extubation delay was also associated with higher mortality (RR for death, 2.2: 95% CI, 1.0 to 4.7).
Patients who died in the hospital were more likely to have delay in extubation and lower admission GCS. Patients with extubation delay had significantly longer ICU length of stay and hospital length of stay and significantly higher charges. (Coplin WM, et al. Am J Respir Crit Care Med 2000;161:1530-1536.)
COMMENT BY UDAY B. NANAVATY, MD
Coplin et al have to be commended for this large prospective cohort study with complex but intriguing statistical analysis. As hypothesized in the study, they show that variation in extubation practice, and especially a delay in extubation once the patient meets accepted criteria for weaning from ventilatory support, is associated with significantly higher morbidity, longer length of stay, and higher charges. More importantly, the study raises a fundamental question that affects every single day of our ICU practice. What are the predictors of successful extubation in patients who satisfy "weaning criteria?"
Weaning criteria simply identify patients who can tolerate spontaneous ventilation, and do not really address the ability of patients to clear airway secretions or their ability to maintain airway patency. It is a simple but important fundamental concept highlighted by this study that successful weaning does not mean successful extubation. This study also raises the concern that there are no standardized criteria regarding readiness for extubation. Clinicians often rely on gag reflex, cough reflex, and a general assessment of secretion burden to decide the readiness of extubation. This study suggests that even more detailed assessment of neurological status or airway function fails to identify the patients who are ready for extubation.
Quite surprisingly, based on this study, one would argue that we often underestimate the patient’s ability to protect their airway or to clear airway secretions. This study suggests that ability to cough spontaneously and the frequency of suctioning may be more reliable indicators of the ability to tolerate extubation. Based on this study it seems likely that the GCS is not a good predictor for ability to tolerate extubation.
One has to be careful when interpreting this prospective observational cohort study since data from some of the patients excluded from this study, especially the patients who underwent tracheotomy prior to extubation, or who died prior to extubation, could have swayed the study results in a significant manner. However, based on this study it seems logical to perform a prospective randomized controlled trial of extubation in a subgroup of stable patients with brain injury to identify factors associated with successful extubation or to identify factors predicting the need for long term airway protection.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content