Association between urea trajectory and protein dose in critically ill adults: a secondary exploratory analysis of the effort protein trial (RE-EFFORT)

Table 2 The estimated effect of time-varying urea and time-varying persistent organ dysfunction on 30-day ICU mortality

	Joint model of time-varying urea HR estimate (95% CrI)	Joint model of time-varying urea and persistent organ dysfunction HR estimate (95% CrI)
Baseline variables
Age	1.29 (1.15–1.45)	1.30 (1.15–1.49)
SOFA	1.09 (0.97–1.23)	1.07 (0.96–1.20)
Randomisation	1.02 (0.74–1.41)	1.03 (0.76–1.41)
RRT	1.06 (0.73–1.57)	1.10 (0.78–1.56)
AKI	1.24 (0.89–1.69)	1.25 (0.91–1.69)
Time-varying variables
Urea	1.34 (1.21–1.48)	1.30 (1.18–1.43)
POD	–	1.32 (1.20–1.45)

Number of subjects: 1277; number of events: 344 (26.9%); number of observations: 11,317. Joint modelling survival analysis allows estimation of the time-varying, patient-specific random effect of the endogenous covariate (urea) on an outcome. Primary and secondary joint models were adjusted for baseline variables (age, renal replacement therapy, sequential organ failure assessment, kidney dysfunction, and protein dose randomisation). Multivariate joint model was adjusted for baseline variables (age, renal replacement therapy, kidney dysfunction, and protein dose randomisation). Effect estimate is for a twofold increase in time-varying urea with 95% credible intervals. Effect estimate for the multivariate joint model is for an increase in one persistent organ dysfunction with 95% credible intervals. The hazard ratios were the adjusted hazard ratios associated with a 1-SD increment in the given variable. Values higher than 1 indicate an increased mortality. The values used for standard deviations were as follows: age, 16.7 years; and SOFA score, 3.7.

ISSN: 1364-8535