Convalescent plasma therapy and mortality in COVID-19 patients admitted to the ICU: a prospective observational study

Annals of Intensive Care

Table 3 Multivariable Cox regression of overall survival according to convalescent plasma therapy in the overall cohort (n = 118)

Multivariable model	Variable	Multivariable hazard ratio	95% CI	p
#1 (n = 118)	CVP	0.39	0.19–0.81	0.011
	paO₂/FiO₂ (per 10 units increase)	0.87	0.80–0.95	0.002
#2 (n = 118)	CVP	0.39	0.18–0.84	0.016
	mSOFA (per 1 point increase)	1.26	1.08–1.46	0.003
#3 (n = 118)	CVP	0.29	0.13–0.66	0.003
	Positivity for SARS-COV-2 antibodies	0.46	0.22–0.95	0.036
#4 (n = 118)	CVP	0.38	0.17–0.85	0.018
	Immunosuppression	1.75	0.65–4.68	0.265
#6 (n = 98)	CVP	0.21	0.09–0.47	< 0.0001
	Positivity for anti-SARS-COV-2 antibodies	0.39	0.18–0.85	0.018
	mSOFA (per 1 point increase)	1.30	1.10–1.54	0.003
	Immunosuppression	1.40	0.51–3.81	0.515

Two patients were excluded because of missing variables. Model#1 is CVP adjusted for paO₂/FiO₂, Model#2 is CVP adjusted for mSOFA-modified sequential organ failure assessment, Model#3 is CVP adjusted for anti-SARS-CoV-2 antibodies positivity at time of ICU admission, Model#4 is CVP adjusted for immunosuppression at time of ICU admission, Model#5 is CVP adjusted for anti-SARS-CoV-2 antibodies positivity at time of ICU admission, mSOFA and immunosuppression In these analyses, the association between CVP and a higher risk of death prevailed after multivariable adjustment