2004 OPEN FORUM Abstracts
ANALYSIS of ventilator Alarms Found out of standard in a children’s hospital
Peter Betit RRT, Michelle Lilley RRT, John
Thompson RRT, John
Arnold MD. Children’s Hospital and Harvard Medical School,
Boston, MA
Background:
Evaluating ventilator alarm function is an important component of
ensuring safe mechanical ventilation. We evaluate ventilator alarm
function as part of a periodic safety check conducted on all
mechanically ventilated patients in 3 ICUs, [neonatal (NICU),
med-surg (MSICU), and cardiac (CICU)], representing 64 beds. All
ventilator alarms are set to a department standard. We report the
analysis of ventilator alarms found out of standard monitored
prospectively over a 5-year period.
Methods: A ventilator
complication report was completed by RTs at the end of each shift.
Ventilator patients, ventilator days, alarms found out of standard
(AOS), and alarms found out of standard / 1000 ventilator hours
(AOS/1K) were recorded and evaluated from 10/99 to 09/03.
Results: There were 10,847 ventilator patients; 1890 NICU, 5284 CICU,
and 3673
MSICU; and 55,122 ventilator days. Overall there were 198.8 ±
51.6 AOS and 1.00 ±
0.37 AOS/1K. The AOS/year were NICU 37.8 ±
12.0, CICU 43.4 ±
3.8, and MSICU 120.2 ±
14.0. The AOS/1K /year were NICU 0.6 ±
0.3, CICU 0.4 ±
0.8, and MSICU 1.3 ± 0.7.
Conclusions: The AOS were highest in the MSICU and lowest
in the NICU. We attribute the differences in AOS between ICUs to the various
patient
populations, and the brand of ventilator used. The relatively low AOS
rate in the NICU may in part be due to the population of pre-term
infants who are typically maintained on minimal ventilator support
until weight targets are attained, and require minimal alarm
adjustments. The downward trend of AOS in the NICU in the last 2
years may be due to the introduction of a new ventilator that
auto-adjusts minute ventilation alarms. The CICU tends to consist of
a more homogenous population of postoperative patients who are weaned
and extubated by protocol in an expeditious manner, and as a result
require fewer alarm adjustments. The MSICU patient population tends
to be less homogeneous with a variety of medical and surgical
diagnoses, and requires more refinements in ventilator modes,
settings and alarms. The patients’physiologic conditions are
more dynamic, and as their condition improves, alarms may become out
of standard between safety checks. The evaluation of AOS is part of
our ongoing ventilator complication QI program and consistent with
JCAHO safety goals. Comparing AOS rates with similar institutions may
help determine an acceptable rate of AOS.