Table 3.
Cut-Off | AUC | Sensitivity (95% CI) |
Specificity (95% CI) |
LR+ | LR− | |
---|---|---|---|---|---|---|
% TCRγδ + | ≥14% | 0.917 | 84.33 (78.80–88.90) |
86.37 (82.67–89.54) |
6.19 | 0.18 |
% sCD3 − | ≤4% | 0.984 | 94.91 (91.07–97.43) |
91.97 (88.91–94.41) |
11.82 | 0.06 |
ratio
TCRγδ + /sCD3 − |
≥5 | 0.992 | 94.91 (91.07–97.43) |
96.11 (93.75–97.76) |
24.38 | 0.05 |
Receiver-operating characteristic curve (ROC) analysis was used to define the best cut-offs points. AUC, area under the curve; CI, confidence interval; and LR, likelihood ratio.