Table 2. Univariate and multivariable associations to 30-day overall survival in 160 septic shock patients.
Univariate Cox models | Multivariable Cox Model 1c (main effects model; N = 152) | Multivariable Cox Model 2d (“interaction” model; N = 152) | ||||||
Prognostic Variablea | Units | Nb | HR (95%CI) | p-value | HR (95%CI) | p-value | HR (95%CI) | p-value |
NFKB1 promoter polymorphism genotype | II | 65 | 1 | 0.03 | 1 | 0.003 | ||
ID/DD | 95 | 1.91 (1.08;3.36) | 2.59 (1.39;4.82) | |||||
Hydrocortison therapy | without HC | 56 104 | 1 | 0.02 | 1 | 0.005 | ||
with HC | 2.15 (1.16;3.98) | 2.64 (1.33;5.23) | ||||||
NFKB1 promoter polymorphism genotype and HC therapy | II with HC | 45 | 1 | <0.0001e | 1 | <0.0001e | ||
II without HC | 20 | 1.03 (0.36;2.97) | 0.74 (0.20;2.70) | |||||
ID/DD with HC | 59 | 3.18 (1.61;6.28) | 3.52 (1.72;7.22) | |||||
ID/DD without HC | 36 | 0.92 (0.37;2.97) | 1.06 (0.42;2.70) | |||||
Agea | per 5 years | 160 | 1.00 (0.92;1.08) | 0.92 | ||||
< 57 years | 77 | 1 | 0.53 | |||||
≥ 57 years | 83 | 0.85 (0.51;1.42) | ||||||
Sex | Female | 59 | 1 | 0.21 | ||||
Male | 101 | 0.71 (0.43;1.20) | ||||||
Height | per cm | 130 | 1.00 (0.98;1.03) | 0.92 | ||||
Body weight | per kg | 131 | 1.00 (0.99;1.02) | 0.49 | ||||
Body-mass-index | per kg/m2 | 129 | 1.01 (0.97;1.06) | 0.55 | ||||
Arterial pressure | per mmHg | 160 | 0.98 (0.97;1.00) | 0.12 | 0.98 (0.96;1.00) | 0.09 | 0.98 (0.96;1.00) | 0.07 |
Noradrenaline dosage | per mg/h | 160 | 1.01 (0.97;1.06) | 0.67 | ||||
Creatinin serum- concentration | per mg/dl | 151 | 1.14 (0.89;1.47) | 0.29 | ||||
Dialysis | no | 57 | 1 | 0.09 | 1 | 0.51 | 1 | 0.54 |
Yes | 95 | 1.65 (0.92;2.95) | 1.22 (0.67;2.23) | 1.21 (0.66;2.20) | ||||
SAPS II | per point | 160 | 1.01 (1.00;1.03) | 0.18 | ||||
SOFA | per point | 111 | 1.06 (0.98;1.15) | 0.15 | ||||
Pro-calcitonin concentration | per µg/l | 145 | 1.00 (1.00;1.01) | 0.26 | ||||
C-reactive protein concentration | per g/l | 145 | 1.01 (0.98;1.03) | 0.56 | ||||
Blood leukocyte concentration | per n/l | 156 | 1.00 (0.98;1.03) | 0.88 |
Multivariable Cox regression analysis for continuous prognostic variables are displayed for the continuous linear predictor; for “Age” we performed several sensitivity analyses including those displayed, in addition to a combined linear and quadratic continuous predictor – these transformations had no impact on the conclusions.
the number of available data for a particular variable in the univariate analysis.
model in which all main effects of potential prognostic factors with univariate uncorrected p-values < 0.15 are included.
like model 1 (see c) but main effects of NFKB1 genotype and hydrocortisone therapy are included as four subgroup.