Skip to main content

Table 2 Key features of the search protocol used to build the final population kinetic model. The strategy is to reduce −2(LL) by more than 3.8 points in each step, which means that the change of the model is statistically significant

From: Distribution of crystalloid fluid infused during onset of anesthesia-induced hypotension: a retrospective population kinetic analysis

Optimization routine

Model

Target parameter

LL

−2 (LL)

AIC

Naive pooled

5 parameters (Vc, k12, k21, k10, and kb)

 

340

−681

−631

Naive pooled

Removal of first fixed parameter

kb

341

−681

−669

FOCE ELS

Removal of first fixed parameter

kb

1280

−2560

−2540

FOCE ELS

Removal of second fixed parameter

k21

1541

−3081

−3065

FOCE ELS

Add covariate: body weight

Vc

1558

−3117

−3099

 

Add covariate: crude MAP

k12

1605

−3210

−3190

FOCE ELS

Add covariate: crude MAP

k10

1632

−3264

−3242

FOCE ELS

Full block model

All the above

1652

−3304

−3276

FOCE ELS

Bootstrap analysis

All the above

1652

−3304

−3246

  1. FOCE ELS forward conditional extended least squares method, LL log likelihood, AIC Akaike criterion