Key characteristics impacting survival of COVID-19 extracorporeal membrane oxygenation

Herrmann, Johannes; Lotz, Christopher; Karagiannidis, Christian; Weber-Carstens, Steffen; Kluge, Stefan; Putensen, Christian; Wehrfritz, Andreas; Schmidt, Karsten; Ellerkmann, Richard K.; Oswald, Daniel; Lotz, Gösta; Zotzmann, Viviane; Moerer, Onnen; Kühn, Christian; Kochanek, Matthias; Muellenbach, Ralf; Gaertner, Matthias; Fichtner, Falk; Brettner, Florian; Findeisen, Michael; Heim, Markus; Lahmer, Tobias; Rosenow, Felix; Haake, Nils; Lepper, Philipp M.; Rosenberger, Peter; Braune, Stephan; Kohls, Mirjam; Heuschmann, Peter; Meybohm, Patrick

doi:10.1186/s13054-022-04053-6

Critical Care

Table 3 Blockwise logistic regression

From: Key characteristics impacting survival of COVID-19 extracorporeal membrane oxygenation

Blocks adjusted		Block 1		Block 1–2		Block 1–3		Block 1–4
Variable	Units	OR	CI 95	OR	CI 95	OR	CI 95	OR	CI 95
1. Demographics, risk factors, comorbidities
Age [years]	19–40	Ref		Ref		Ref		Ref
	41–70	3.11	[1.71;5.78]	2.75	[1.49;5.15]	2.71	[1.46;5.10]	2.48	[1.32;4.71]
	> 70	12.31	[4.07;46.66]	8.70	[2.78;33.78]	8.28	[2.63;32.25]	6.81	[2.13;26.90]
	Missing	2,329,070.78	[0.00;NA]	2,976,862.44	[0.00;NA]	3,313,285.99	[0.00;NA]	1,813,873.83	[0.00;NA]
Sex	m	Ref		Ref		Ref		Ref
	w	0.87	[0.57;1.34]	0.79	[0.51;1.24]	0.78	[0.50;1.23]	0.87	[0.55;1.38]
	Missing	0.89	[0.18;6.34]	1.06	[0.22;7.66]	0.97	[0.20;7.03]	1.12	[0.21;8.39]
BMI [kg/m²]	< 25	Ref		Ref		Ref		Ref
	25–30	0.63	[0.35;1.13]	0.65	[0.35;1.17]	0.65	[0.35;1.18]	0.62	[0.33;1.13]
	30–35	0.70	[0.37;1.32]	0.71	[0.37;1.35]	0.73	[0.38;1.40]	0.74	[0.38;1.43]
	≥ 35	0.54	[0.29;0.99]	0.57	[0.30;1.07]	0.57	[0.30;1.06]	0.51	[0.26;0.97]
	Missing	2.49	[0.74;11.48]	3.16	[0.91;14.85]	3.13	[0.89;14.84]	3.25	[0.90;15.77]
Immunosuppression within 6 months prior to admission	No	Ref		Ref		Ref		Ref
	Yes	1.82	[0.84;4.44]	1.69	[0.76;4.16]	1.75	[0.78;4.34]	1.44	[0.63;3.63]
	Unknown	0.61	[0.39;0.95]	0.73	[0.42;1.31]	0.76	[0.42;1.36]	0.77	[0.41;1.44]
2. Severity of disease
Intubation prior to ECMO [days]	< 5			Ref		Ref		Ref
	5–7			2.67	[1.54;4.81]	2.65	[1.52;4.81]	2.39	[1.35;4.37]
	≥ 8			1.31	[0.79;2.18]	1.26	[0.75;2.10]	1.30	[0.76;2.22]
	No prior intubation			0.88	[0.45;1.73]	1.12	[0.55;2.33]	1.90	[0.87;4.25]
	Missing			1.57	[0.76;3.43]	1.86	[0.88;4.13]	1.93	[0.89;4.42]
EOLIA criteria	Liberal			Ref		Ref		Ref
EOLIA criteria	Restrictive			0.62	[0.40;0.95]	0.62	[0.40;0.95]	0.64	[0.41;0.99]
3. ECMO Case volume	Low					Ref		Ref
	Medium					0.81	[0.44;1.44]	0.79	[0.42;1.43]
	High					0.53	[0.28;0.97]	0.55	[0.28;1.02]
4. Complications
Major bleeding or thromboembolic event	No							Ref
Major bleeding or thromboembolic event	Yes							1.70	[1.12;2.57]
Secondary bacterial infection (respiratory tract or bloodstream)	No							Ref
	Yes							0.75	[0.48;1.16]
Renal replacement therapy	No							Ref
Renal replacement therapy	Yes							2.35	[1.60;3.46]
AIC		815.24		804.94		803.01		780.89

Variables associated with mortality during ICU stay and odds ratios (ORs) with corresponding 95% confidence intervals (CIs) in blockwise logistic regression. Variables were selected a priori based on clinical background and assigned to blocks reflecting the clinical course over time. References for each variable are indicated (ref). Models were adjusted in four blocks: 1. Demographics, Risk factors and Comorbidities; 2. Severity of disease; 3. ECMO case volume, and 4. Complications. Step-wise integration of blocks into the model is depicted in the table from the left (block 1 only) to the right (block 1–4). The quality of the models was assessed using the Akaike information criterion (AIC). Statistical significance was determined at an α level of 0.05 (two-tailed)

Back to article page

ISSN: 1364-8535

Contact us

Submission enquiries: journalsubmissions@springernature.com