Monitoring of circulating monocyte HLA-DR expression in a large cohort of intensive care patients: relation with secondary infections

Annals of Intensive Care

Table 2 Multivariate cause-specific survival model: sensitivity analysis of predictors of secondary infection occurrence

Variable	Cs-HR	IC95%	p
Log mHLA-DR1	0.90	[0.67; 1.19]	0.46
Low mHLA-DR 1	1.06	[0.72; 1.57]	0.77
Log mHLA-DR2	0.63	[0.48; 0.81]	0.00
Low mHLA-DR2	1.75	[1.21; 2.53]	0.00
Slope (%)	0.98	[0.97; 0.99]	0.00
Slope (↘)	1.61	[1.13; 2.3]	0.01
Low mHLA-D1 and down slope	1.95	[1.17; 3.26]	0.01
Lymphopenia 1	1.62	[1.13; 2.32]	0.01
Lymphopenia 2	0.87	[0.58; 1.3]	0.51
Lymphopenia down	0.98	[0.69; 1.41]	0.93
Monocytopenia 1	0.94	[0.65; 1.34]	0.71
Monocytopenia 2	0.59	[0.41; 0.85]	0.00
Monocytopenia down	0.66	[0.47; 0.92]	0.02

The cause-specific model is a common alternative to survival analysis for handling competing risks. This model allows a quantification of the cause-specific relative hazard, which is the association between the exposure and the outcome when the individuals with the competing event are censored
The following covariates were used: comorbidities (without immunosuppression), immunosuppression, SOFA score and motif of admission
The competing event was “discharged alive before 28 days” and the outcome was nosocomial infection