Risk factors for unplanned extubations in critically ill patients, using PRISMA analysis

Introduction

Unplanned extubations (UEs) occur in 4.2–14% of critically ill patients on mechanical ventilation. UE is defined as premature removal of endotracheal tube. UEs are associated with increased morbidity, mortality and utilisation of resources. We collected data from patients who extubated themselves during a 6-month period. The aim of the study was to assess risk factors for UE using the Prevention and Recovery Information System for Monitoring and Analysis (PRISMA) method. The method is used in the chemical and steel industry, and nowadays in health care. The goal of this method is to build a quantitative database of incidents and process deviations from which conclusions may be drawn to suggest preventive measures. Incidents are described by means of causal trees to identify root causes.

Method

In a 28-bed surgical, thoracic surgical and medical ICU 24/7 all UEs were directly reported by telephone to the main investigator. By means of a structured interview, data on the specific circumstances of the UE were obtained from the nurse and doctor involved. By exploring the situational characteristics of a UE by interviewing medical and nursing staff, we developed in collaboration with professionals in Patient Safety Systems a causal tree consisting of three main contributing actions: inadequate treatment agitation, patient nonacceptance of tube, and no prevention of UE. Each UE was analysed by two investigators independently to identify which of the main actions contributed. Combinations of causes were subsequently identified, leading to identification of root causes.

Results

Twenty-five patients were enrolled with 640 ventilation periods and 2962 ventilation days. This yielded an incidence of 3.9% and 0.8 UEs per 100 ventilation days. The mean age was 58.2 ± 17.0. Male–female ratio was 2.1. The mean APACHE score was 16.6 ± 5.0. At the time of UE, the mean Ramsay sedation score was 1.88 ± 0.7 and the weaning–nonweaning ratio 1.8. Reintubation was needed in 48% of the cases.

In 32% of the patients, inadequate treatment agitation led to UE. In the case of patient nonacceptance of tube (56%), 57% patients were in delirium and 43% were fully conscious left unnecessary intubated. In 83% this delay in extubation despite good clinical condition was due to suboptimal management. In 16%, the decision to extubate the patient was not executed because of the night hour. No prevention of the actual extubation was in 64% related to no detection that occurred because personnel was busy elsewhere (80%) and/or not expecting the extubation (94%). No expectation coincided with false assessment of adequacy of the fixation technique (92%).

Conclusion

Incident analysis by the PRISMA method is a feasible method to identify contributing factors to UE. Optimalisation of agitation treatment and extubation protocols, and increased awareness of high-risk conditions, could prevent UEs.