Table 3.

ROC curve analyses

Variable	AUC	SE	95%CI	threshold values	sensitivity	95%CI	specificity	95%CI	positive likelihood ratios	95%CI	negative likelihood ratios	95%CI
BEa	0.693	0.0143	0.675–0.712	≤ − 4.6	59.19	54.7–63.6	71.84	69.8–73.8	2.10	1.9–2.3	0.57	0.5–0.6
Lactateb	0.715	0.0132	0.697–0.733	> 2.42	69.49	65.2–73.5	60.74	58.5–62.9	1.77	1.6–1.9	0.50	0.4–0.6
pHc	0.670	0.0149	0.651–0.689	≤7.24	41.82	37.4–46.3	87.00	85.4–88.5	3.22	2.8–3.8	0.67	0.6–0.7

Pairwise comparison of AUCs: a vs b: P = 0.068; a vs c: P = 0.020; b vs c: P < 0.001.

AUC: area under the curve; SE: standard error; CI: confidence interval.