Skip to main content
. 2021 Jun 30;12:658843. doi: 10.3389/fimmu.2021.658843

Table 2.

The performance of different methods for distinguishing between ATB and LTBI.

Methods Cutoff value AUC (95% CI) Sensitivity (95% CI) Specificity (95% CI) PPV (95% CI) NPV (95% CI) PLR (95% CI) NLR (95% CI) Accuracy
CD4+ T cell number (/μl) 404 0.788 (0.694-0.882) 57.41% (44.16%-69.67%) 91.23% (81.06%-96.20%) 86.11% (71.34%-93.92%) 69.33% (58.17%-78.61%) 6.54 (2.75-15.59) 0.47 (0.34-0.64) 74.77%
CD8+ T cell number (/μl) 203 0.633 (0.528-0.737) 35.19% (23.82%-48.52%) 85.96% (74.68%-92.71%) 70.37% (51.52%-84.15%) 58.33% (47.65%-68.29%) 2.51 (1.2-5.24) 0.75 (0.6-0.94) 61.26%
NK cell number (/μl) 156 0.852* (0.778-0.927) 68.52% (55.26%-79.32%) 87.72% (76.75%-93.92%) 84.09% (70.63%-92.07%) 74.63% (63.07%-83.51%) 5.58 (2.72-11.43) 0.36 (0.24-0.54) 78.38%
B cell number (/μl) 93 0.629 (0.523-0.736) 38.89% (27.04%-52.21%) 85.96% (74.68%-92.71%) 72.41% (54.28%-85.30%) 59.76% (48.94%-69.70%) 2.77 (1.34-5.72) 0.71 (0.56-0.9) 63.06%
HLA-DR+CD3+ T cells (%) 24.7 0.611 (0.504-0.719) 40.74% (28.68%-54.03%) 80.70% (68.66%-88.87%) 66.67% (49.61%-80.25%) 58.97% (47.89%-69.22%) 2.11 (1.13-3.93) 0.73 (0.57-0.95) 61.26%
Treg (%) 3.82 0.613 (0.506-0.720) 44.44% (32.00%-57.62%) 80.70% (68.66%-88.87%) 68.57% (52.02%-81.45%) 60.53% (49.29%-70.75%) 2.3 (1.25-4.23) 0.69 (0.53-0.9) 63.06%
CD4+ T cell function (%) 13.8 0.766 (0.678-0.854) 42.59% (30.33%-55.84%) 92.98% (83.30%-97.24%) 85.19% (67.52%-94.09%) 63.10% (52.42%-72.63%) 6.07 (2.25-16.41) 0.62 (0.49-0.79) 68.47%
CD8+ T cell function (%) 41.2 0.782 (0.692-0.873) 62.96% (49.63%-74.58%) 92.98% (83.30%-97.24%) 89.47% (75.87%-95.83%) 72.60% (61.44%-81.51%) 8.97 (3.41-23.59) 0.4 (0.28-0.57) 78.38%
NK cell function (%) 62.1 0.744 (0.650-0.838) 37.04% (25.42%-50.37%) 91.23% (81.06%-96.20%) 80.00% (60.87%-91.14%) 60.47% (49.90%-70.14%) 4.22 (1.71-10.45) 0.69 (0.55-0.86) 64.86%
Diagnostic model 0.676 0.920 (0.867-0.973) 81.48% (69.16%-89.62%) 91.23% (81.06%-96.20%) 89.80% (78.24%-95.56%) 83.87% (72.79%-91.00%) 9.29 (3.98-21.66) 0.2 (0.12-0.36) 86.49%

*Compared with diagnostic model using z statistic, P < 0.05; compared with diagnostic model using z statistic, P < 0.01; compared with diagnostic model using z statistic, P < 0.001; ATB, active tuberculosis; LTBI, latent tuberculosis infection; AUC, area under the curve; PPV, positive predictive value; NPV, negative predictive value; PLR, positive likelihood ratio; NLR, negative likelihood ratio; CI, confidence interval.