Abstract
Background:
Tracheostomies are performed for a variety of indications, predominantly to facilitate weaning from invasive mechanical ventilation or to establish a stable, permanent artificial airway. If a patient is deemed ready to be assessed for liberation intermittent trials of unassisted breathing (trach collar trials [TCTs]) are performed. In our institution TCTs are not standardized and are usually terminated by individual clinicians based on either physiologic parameters or after a pre-defined time limit is reached. The purpose of this quality improvement project is to review the consequences of our TCT clinical practice.
Methods:
As part of an IRB exempt, departmental quality improvement project, data on all adult tracheostomy patients undergoing unassisted breathing trials within our five adults ICUs were recorded between August 2015 and September 2016. TCT details were recorded prospectively each day by respiratory therapists. Included was: reason for tracheostomy, time from mechanical ventilation to tracheostomy, time from tracheostomy to liberation, ventilator settings, length of initial trial of unassisted breathing, and reason for trial termination. Subjects with a pre-existing tracheostomy were excluded from data on time to tracheostomy, and time from tracheostomy to liberation. Subjects who were not liberated were excluded from total days of mechanical ventilation. Data were analyzed using SPSS v24 and described as median (interquartile range) and percentages. Nonparametric testing was performed to compare differences between ICUs due to non-normally distributed data.
Results:
A total of 190 patients were evaluated. 19 (10%) of subjects were not liberated from mechanical ventilation and 23 (12%) subjects had tracheostomy in place when mechanical ventilation was initiated. Data are summarized in Table 1. Time to tracheostomy (P=0.008), days from tracheostomy to liberation (P=0.001), inspiratory pressure/pressure support (P<0.001) varied significantly between different ICUs. There were no statistically significant differences for total days of mechanical ventilation, total trials, PEEP, FIO2, or length of first trial between different ICUs. Reason for trial termination varied significantly (P<0.001).
Conclusions:
Clinical practice appeared to vary significantly between units with half of the unassisted breathing trials being terminated due to a time goal or per team decision and 43% being terminated due to physiologic reasons.
*=excludes subjects with existing tracheostomy, +=excludes subjects not liberated, IP=inspiratory pressure, PS=pressure support, MV=mechanical ventilation View all access options for this article.Table 1 - Results for all subjects, n=191
Median (IQR)
Days to Tracheostomy, n=172*
10 (6-15)
Days from Tracheostomy to Liberation from MV, n=167*
9 (4-19)
Total Days of MV, n=171+
20 (13-32)
Total Trials
4 (2-7)
Trial number when liberated
4 (2-7)
IP/PS, cmH2O
8 (5-10)
PEEP, cmH2O
8 (5-8)
FiO2
0.30 (0.30-35)
TCT Time (first trial only), hours, n=171
5.6 (2.3-11.1)
Reasons for TCT Termination
n, (%)
Total TCTs
1301
TCTs stopped
1131
Time goal achieved/per team
561 (50%)
Physiology
483 (43%)
Procedure
29 (3%)
Not Recorded
58 (5%)
Get full access to this article