Table 2.
Prediction of seroreactivities of Chlamydia trachomatis antigen-based antibody detection.a
| C. trachomatis antigen | Sensitivity (%) based on different specificityb | p value c | AUC ± SE e | p valuef | |||
|---|---|---|---|---|---|---|---|
| Specificity = 90% | Specificity = 95% | Specificity = 98% | Average | ||||
| Mikrogen | 97.9 | 97.0 | 95.5 | 96.8 | Ref d | 0.993 ± 0.0031 | Ref |
| Pgp3 | 96.8 | 95.7 | 94.1 | 95.5 | 0.585 | 0.986 ± 0.0056 | 0.207 |
| TmeA | 94.3 | 91.9 | 88.5 | 91.6 | 0.041 | 0.976 ± 0.0075 | 0.006 |
| InaC | 94.2 | 92.5 | 90.2 | 92.3 | 0.072 | 0.969 ± 0.0107 | 0.005 |
| HSP60 | 94.3 | 90.8 | 85.1 | 90.0 | 0.013 | 0.976 ± 0.0095 | 0.065 |
a125 Chlamydia trachomatis NAAT positive and 125 C. trachomatis negative sera from low exposure risk children were used to evaluate the performance of different C. trachomatis antigen-based antibody detection.
bThe sensitivities of each assays at different specificity cutoff values were calculated by ROC analysis.
cComparisons of the average sensitivity was carried out by Student’s t-test.
dMikrogen-CT ELISA was set as reference.
eArea under ROC curve, an AUC value of 0.50 indicates that discrimination of positive from negative data equals random categorization (no discrimination), and an AUC value of 1.0 indicates that prediction is perfect (100% accurate discrimination). SE, standard error.
fComparisons of AUC between assays was carried out by Delong test.