Skip to main content
. 2022 Feb 17;17(2):e0262938. doi: 10.1371/journal.pone.0262938

Table 3. ROC curve analysis assessing the various biomarkers among patients with bacteremic infection compared to those with no evidence of bacterial infection.

Parameter AUC 95% CI p-value for AUC LR Sensitivity Specificity Cut off value NPV PPV
(%) (%) (%) (%)
nCD64: lCD64 0.86 0.75–0.98 0.0003 4.25 92.3 78.3 > 7.4 70.6 94.7
nCD64: mHLA-DR 0.95 0.88–1.02 < 0.0001 19.5 84.6 95.7 > 0.34 90.9 88.0
NE-WY 0.78 0.63–0.93 0.005 3.54 61.5 82.6 > 750.5 66.7 79.2
NE-SFL 0.84 0.70–0.98 0.0007 4.87 84.6 82.6 > 49.75 73.3 90.5
Automated IG% 0.78 0.62–0.93 0.006 2.65 69.2 73.9 > 0.85 60.0 81.0
Abs auto IG 0.74 0.57–0.91 0.017 3.98 69.2 82.6 > 0.065 69.2 82.6
CRP + 1 1.00–1.00 0.0002 >17 100 100 > 12.5 100.00 95.0
PCT * 1 1.00–1.00 0.037 >8 100 100 >0.48 100.0 100.0

+ N = 24

*N = 10

AUC, area under the curve; CI, confidence interval; LR, likelihood ratio; NPV, negative predictive value; PPV, positive predictive value; nCD64:lCD64, neutrophil CD64:lymphocyte CD64; nCD64:mHLA-DR, neutrophil CD64:monocyte HLA-DR; NE-WY, fluorescent light distribution width of the neutrophil area; NE-SFL, fluorescent light intensity of the neutrophil area; IG%, immature granulocyte percentage; CRP, C-reactive protein; Abs auto IG, absolute automated IG count.