Extracorporeal membrane oxygenation in non-intubated immunocompromised patients

Table 2 Predictors for failure of the awake ECMO concept

Characteristic	Secondary intubation			Logistic regression
Characteristic	No	Yes	p	OR	95%-CI	p
Benzodiazepine use during ECMO support—no (%)	2/7 (28.6)	8/11 (72.7)	0.066	6.7	0.8–55	0.078
Ppeak (NIV) before ECMO initiation—cmH2O	15 (11–16)	19 (17–22)	0.014	1.6	1–2.6	0.05
pCO2 before ECMO initiation—mmHg	37 (35–43)	50 (34–76)	0.06	1.1	1–1.2	0.163
ECMO support duration—days	9 (8–11)	12 (10–28)	0.049	1.2	0.9–1.5	0.173

Description of parameters that were associated with the necessity of later secondary intubation. Factors associated with later failure of an awake ECMO concept were more prominent use of benzodiazepines during awake ECMO support, higher peak pressures applied in noninvasive ventilation and hypercapnia directly before ECMO insertion as well as longer ECMO support. Values are presented as median (25–75% interquartile range) or if categorical as numbers and percentage
CI, Confidence Interval; NIV, Noninvasive Ventilation; Ppeak, Peak Pressure

ISSN: 1364-8535