Table 2.

Multivariate mixed-effects model for predictors of ΔECV

VariableUnivariate ΔECVMultivariate ΔECVMultivariate ΔECV: variable × time
r∗ 95% CIP valueEstimate95% CIP value
Δ cardiac index 0.4†  −0.03 to 0.7 .06    ns 
ΔLV mass index 0.37 −0.05 to 0.68 .09 −0.48 −0.78 to 0.09 .004 ns 
Δ reticulocyte count 0.36 −0.05 to 0.66 .09    ns 
ΔeGFR 0.41†  −0.01 to 0.7 .05    ns 
ΔECM volume index 0.9 0.78-0.96 <.001 1.47 0.49-1.78 <.001 ns 
VariableUnivariate ΔECVMultivariate ΔECVMultivariate ΔECV: variable × time
r∗ 95% CIP valueEstimate95% CIP value
Δ cardiac index 0.4†  −0.03 to 0.7 .06    ns 
ΔLV mass index 0.37 −0.05 to 0.68 .09 −0.48 −0.78 to 0.09 .004 ns 
Δ reticulocyte count 0.36 −0.05 to 0.66 .09    ns 
ΔeGFR 0.41†  −0.01 to 0.7 .05    ns 
ΔECM volume index 0.9 0.78-0.96 <.001 1.47 0.49-1.78 <.001 ns 

A multivariate linear mixed-effects model was constructed with random intercepts and random slopes for time at the patient level. Fixed effects included time, person-mean, and within-patient components for all variables with P value <.1 in univariate analyses. The outcome was ΔECV. Predictors were mean centered to facilitate interpretation of main effects. The model explained 91% of the variance (R2 = 0.91).

CI, confidence interval; ECM, extracellular matrix; ns, not significant.

∗

Pearson correlation reported unless otherwise indicated.

†

Spearman rank correlation.

View Large
This Feature Is Available To Subscribers Only

or Create an Account

Close Modal
Close Modal