| Description |
Problem/Background: Extubation failure (EF) is the need for reintubation within 48 hours after a planned extubation in patients meeting weaning criteria from mechanical ventilation (MV). The incidence rate varies in the literature but has a statistically significant correlation with mortality, prolongation of MV, and increased the length of stay in the intensive care unit and hospital. Causes, contributory factors, and predictive models have been extensively studied with mixed results. Weaning from MV and extubation are essential during the hospitalization of critically ill patients, and failure is considered an adverse event. The use of non-invasive ventilation (NIV) and high flow nasal cannula (HFNC) oxygen therapy during the post-extubation period has demonstrated effectiveness for a subset of patients considered high-risk. The healthcare institution does not measure EF, and the practice of prophylactic NIV/HFNC has provider variability. Methods: This quality improvement initiative aimed to assess readiness for extubation planning within a 16-bed critical care unit. Initial data obtained was a two-year retrospective review of patient characteristics and EF incidence. An extubation failure risk assessment (EFRA) was assembled using evidence from 32 high-quality studies and clinical practice guidelines. The EFRA was used by nursing staff during a five-month implementation phase. Data was collected from the EFRA and medical record to identify high-risk patients and determine the rate of best practice deviation. An evaluation survey was sent to the nursing staff after implementation to determine the usability of the assessment and feasibility in its continual use. Results: The baseline assessment identified an EF rate of 8.9% (N=13) during the two previous years. A large proportion of patients (50.7%, N=150) were excluded due to intubation indication (n=43), transition to comfort measures only (n=57), and tracheostomy placement (n=36). Extubation failure had a positive statistical correlation with inpatient mortality (p<0.001), but no statistical correlation with any particular risk factor (medical admission diagnosis, history of chronic obstructive pulmonary disease, heart failure, body mass index, sepsis). During the ERAT integration phase, there was a low EF rate (8.1%, n=3) despite a large patient proportion identified as high-risk (64.9%, n=24). Of the patients identified as high-risk, few received prophylactic NIV (16.67%, n=4) or HFNC (12.5%, n=3). The calculated odds ratio for a high-risk patient to experience EF was no different than one considered low risk (OR=1.091 [95% CI 0.089 to 13.31; p=0.005]). Survey responses were significant for the usability, complexity, ease, and confidence in use for the ERAT with favor towards electronic integration. The average amount of time spent away from patient care to complete an assessment was 4.73 + 3.13 minutes. Conclusion: Findings revealed a low EF rate and no particular risk factor with a strong statistical correlation. Despite a large proportion of patients identified as high-risk, the EF rate remained low despite a high rate of best practices deviation. This project is unique in that it attempted to utilize best practice guidelines regarding EF into routine patient care and added transparency to a previously unmeasured event. |
