Effects of Intubation Location on Risk of Ventilator Associated Events

Margaret Lind | 2017

Advisor: Stephen E. Hawes

Research Area(s): Clinical Epidemiology



Ventilation is a crucial tool in modern medicine that allows for advanced respiratory support in critically ill patients. However, prolonged ventilation has been found to be linked with ventilator-associated conditions (VACs), which in turn are associated with multiple adverse hospital outcomes.


To evaluate the association between intubation practices (using intubation locations as a proxy) on the risk of VACs, ventilator-associated conditions (VACs), infection-related ventilator-associated complications (IVACs), and possible VAPs (PVAPs) and the mediating effect of intubation location on the association between VACs and adverse hospital outcomes including prolonged hospital length of stay, prolonged ventilation duration, and hospital mortality.


We conducted a retrospective cohort study on all patients who underwent mechanical ventilation for two or more days at Harborview Medical Center (HMC), a 413-bed Trauma 1 Center and teaching hospital in Seattle, Washington. Data was collected between June 2015 and November 2016.


Of the 3,424 patients who were ventilated during our study period, 1,323 had a ventilation duration of 2 or more days and were included in our study. Non-communicable diagnosis was the primary reason for hospitalization (49.06%), and patients tended to be male (70.98%). ICU and in-field intubations were associated with a non-significantly lower hazard of total VACs and PVAPs than emergency room intubations. The hazard of PVAPs was 0.29 (CI: 0.04 – 2.18; P= 0.23) times lower in individuals intubated in the ICU and 0.30 (CI: 0.08 – 1.17; P = 0.08) times lower in individuals intubated in the field than the individuals intubated in the emergency room while holding confounding factors constant. The hazard of IVACs was non-significantly higher in individuals intubated in the ICU (HR: 1.71; CI: 0.383 – 7.59; P = 0.48) and in the field (HR: 1.68; CI: 0.57 – 4.97; P = 0.35) than individuals intubated in the emergency room. Intubation locations did not significantly modify the associations between VAC and any of the outcomes of interest.


Although we did not find a significant associated between intubation location and VAC, non-infectious VACs, IVACs, and PVAPs, the magnitude of the point estimates justifies further research on this topic. We recommend that additional analyses are conducted to examine the directionality of the association between VACs and ventilation duration.