Can a Risk Assessment Tool Get Everyone on the Same Page?*
The evidence for harm in unplanned extubation events is compelling. The prevalence of cardiovascular collapse following unplanned extubation has been reported to be as high as 20%, with infants and neonates at greatest risk (2). Roddy et al (3) found that the hospital length of stay in patients who had an unplanned extubation event was increased by 6.5 days when compared with age and diagnosis-matched controls. Furthermore, hospital costs were increased, on average, by $36,000 for patients who had an unplanned extubation event (3).
In this issue of Pediatric Critical Care Medicine, Vats et al (4) propose an airway risk assessment score for unplanned extubation in pediatric patients. As part of a system-wide quality improvement initiative, their objective was to produce a standardized airway assessment tool with the potential to stratify patients by risk for unplanned extubation.
The risk assessment tool proposed by Vats et al (4) has many strengths. The tool adheres to several best practices recommendations, for which there is variable evidence (5). The frequency of assessments (at 04:00 and 16:00 each day) acknowledges the dynamic nature of unplanned extubation risk, accounting for higher risk activities like bedside procedures and patient transport (5). Review of the score on rounds contributes to improved situational awareness among the multidisciplinary care team, giving the team valuable opportunities to address aspects of an individual patient’s risk score. The uniformity of the tool for all ICUs—neonatal, pediatric, and cardiac—should serve to improve scoring accuracy, score interpretation, and compliance with the tool since two of the major stakeholder groups, respiratory therapists and bedside nurses, often float between units.
There are also, however, several limitations to the evaluation of the validity of the risk assessment tool, many of which are acknowledged by the authors. Nearly one quarter of patients did not receive a risk score, though given the overwhelming majority of patients in the missing values groups with mechanical ventilation less than 48 hours, one might presume that these were largely low-risk patients. If we accept this presumption, it would suggest that both the frequency and events per 100 ventilator days are overestimated for low-risk patients, which further distinguishes the low-risk group from all of the higher risk strata. The high point value for anatomical risk (difficult airway and/or facial deformities) seems to conflate unplanned extubation risk with the risk of catastrophic airway emergency should unplanned extubation occur.
Among patients with unplanned extubation, the authors found that 38% of these patients did not require reintubation. This rate falls right of the middle of published ranges for reintubation following unplanned extubation and highlights the need for daily assessments of extubation readiness (5). Although the clinical (and statistical) significance is not entirely clear, review of the supplemental digital content (Figure 1, A–H) suggests that patients who required reintubation may have increased agitation and oral secretions when compared with those who did not require reintubation. It would be interesting to see if trends develop over time using the tool that helps discriminate these two groups and potentially lead to earlier extubation. This is yet another strength of using an uniform tool among a number of ICUs.
Quite possibly, the biggest confounder in the prevention of unplanned extubation relates to the use of sedative and hypnotic agents.