Objectives: This study aimed to explore the
relationship between the variables of mechanical ventilation and circulatory
perfusion and its association with ICU mortality during the first day of
mechanical ventilation.
Design: Retrospective cohort study.
Setting: The Department of Critical
Care Medicine, Peking Union Medical College Hospital. Patients: Patients who
have undergone mechanical ventilation. Interventions: None. Measurements and
Main Results: This study used the
main clinical data obtained from the real-time bedside messaging systems of
mechanically ventilated patients during their first day in the ICU from May
2013 to May 2016, including data on the variables of mechanical ventilation and
circulatory perfusion. An analysis was then performed on the association of the
above data with the patient’s in-ICU mortality. There were 5,103 patients who
received mechanical ventilation during this period, and of these, 309 patients
died during their ICU treatment. Peak airway pressure, mean airway pressure,
respiratory rate, heart rate, mean arterial pressure, FIO2, blood oxygen
saturation, PO2, peripheral perfusion index, and lactate level were correlated
with patient outcomes. A Cox logistic regression analysis suggested that mean
airway pressure and perfusion index were the most independent risk and
protective factors, respectively, for patient ICU mortality. The areas under
the curve for a poor prognosis for mean airway pressure and perfusion index
were 0.799 (95% CI, 0.77–0.829) and 0.759 (95% CI, 0.729–0.789), respectively.
Further, mean airway pressure and perfusion index exhibited a causal
interaction. The relative excess risk due to interaction was 2.061 (–0.691 to
4.814), the attributable proportion due to interaction was 0.210 (–0.027 to
0.447), and the synergy index was 1.306 (0.930–1.833).
Conclusions: A higher mean airway pressure
and lower perfusion index provided a worse prognosis in mechanically ventilated
patients, and it appears that these two variables have a casual interaction.
No comments:
Post a Comment