Skip to main content

Table 5 A multivariate Cox proportional hazards model with time-dependent variables for the independent factors of long-term mortality (N = 53,833)

From: Kidney function decline after a non-dialysis-requiring acute kidney injury is associated with higher long-term mortality in critically ill survivors

Covariate HR (95% CI) P
Demographic data
organ transplantation 0.29 (0.09, 0.98) 0.05
age 1.04 (1.02, 1.05) < 0.001
ICU admission and surgery
CPR*ECMO* 0.17 (0.03, 0.99) 0.048
MAP at the peak of AKI 0.99 (0.97, 1.00) 0.006
general surgical services 1.77 (1.19, 2.64) 0.005
receive CPR 6.55 (2.34, 18.32) < 0.001
Time-dependent variables during follow-up*
preserved eGFR during follow-up 1 reference
baseline eGFR*CKD3-TD 1.007 (1.001, 1.012) 0.016
baseline eGFR*CKD4-TD 1.016 (1.010, 1.021) < 0.001
baseline eGFR*CKD5-TD 1.017 (1.008, 1.026) < 0.001
baseline eGFR*ESRD-TD 1.054 (0.959, 1.158) 0.3
  1. AKI, acute kidney injury; CI, confidence interval; CKD, chronic kidney disease; CPR, cardiopulmonary resuscitation; ECMO, extracorporeal membrane oxygenation; eGFR, estimated glomerular filtration rate; ESRD, end-stage renal disease; HR, hazard ratio; MAP, mean arterial pressure. *CPR*ECMO indicates the interaction term of receiving both CPR and ECMO; Baseline eGFR*CKD3 (4, 5)-TD indicates the interaction term of baseline eGFR and stage 3 (4, 5) CKD time-dependent variables; Baseline eGFR*ESRD-TD indicates the interaction term of baseline eGFR and ESRD. The main CKD3, 4, 5-TD factors were not significant in the final model.