Table 1.
Overall diagnostic performance of the different assays
| N | Roche Ig anti-N |
Abbott IgG anti-N | Euro NCP IgG anti-N | Mikrogen IgG anti-N |
Maglumi IgG anti-N/S | Diasorin IgG anti-S | Euro S1 IgG anti-S | |
|---|---|---|---|---|---|---|---|---|
| Sensitivity (95% CI) | 223 | 71.8% (65.5–77.3) |
70.9% (64.6–76.4) |
73.1%d (66.9–78.5) |
70.4% (64.1–76.0) |
68.6% (31.8–47.1) |
63.2% (56.7–69.3) |
64.6% (58.1–70.6) |
| D0–6 | 43 | 32.6% (20.4–47.6) |
27.9% (16.6–42.8) |
30.2% (18.5–45.2) |
30.2% (18.5–45.2) |
25.6% (14.8–40.4) |
14.0% (6.2–27.6) |
18.6% (9.5–32.9) |
| D7–13 | 98 | 69.4% (59.7–77.7) |
67.3% (57.5–75.9) |
71.4% (61.8–79.5) |
67.3% (57.5–75.9) |
64.3% (54.4–73.1) |
58.2% (48.3–67.5) |
60.2% (50.3–69.3) |
| D14–17 | 42 | 92.9% (80.3–98.2) |
95.2% (83.3–99.5) |
95.2% (83.3–99.5) |
90.5% (77.4–96.8) |
92.9% (80.3–98.2) |
90.5% (77.4–96.8) |
88.1% (74.5–95.3) |
| D18–21 | 16 | 93.8% (69.7–100) |
100% (77.3–100) |
100% (77.3–100) |
100% (77.3–100) |
100% (77.3–100) |
100% (77.3–100) |
100% (77.3–100) |
| D22–27 | 13 | 100% (73.4–100) |
100% (73.4–100) |
100% (73.4–100) |
100% (73.4–100) |
100% (73.4–100) |
100% (73.4–100) |
100% (73.4–100) |
| D28–37 | 11 | 100% (70.0–100) |
100% (70.0–100) |
100% (70.0–100) |
100% (70.0–100) |
100% (70.0–100) |
100% (70.0–100) |
100% (70.0–100) |
| Specificity (95% CI) | 113 | 100% (96.1–100) |
99.1% (94.7–100) |
94.7% (88.7–97.8) |
96.5% (91.0–98.9) |
96.5% (91.0–98.9) |
99.1% (94.7–100) |
96.5% (91.0–98.9) |
| Other coronavirus | 24 | 100% (83.7–100) |
100% (83.7–100) |
100% (83.7–100) |
100% (83.7–100) |
100% (83.7–100) |
100% (83.7–100) |
100% (83.7–100) |
| Respiratory infection | 49 | 100% (91.3–100) |
100% (91.3–100) |
93.9% (82.9–98.5) |
93.9% (82.9–98.5) |
98.0% (88.3–100) |
100% (91.3–100) |
93.9% (82.9–98.5) |
| Antiviral antibodies | 40 | 100% (89.6–100) |
97.5% (86.0–100) |
92.5% (79.4–98.1) |
97.5% (86.0–100) |
92.5% (79.4–98.1) |
97.5% (86.0–100) |
97.5% (86.0–100) |
| LR+ | +∞ | 80.1 | 13.8 | 19.9 | 19.4 | 74.1 | 18.2 | |
| ROC curve (area) | All | 0.950 | 0.907 | 0.928 | 0.864 | 0.871 | 0.865 | 0.909 |
| Cut-off (Manufacturer)a | 1.0 | 1.4 | 0.8/1.1 | 20/24 | 1.0 | 12/15 | 0.8/1.1 | |
| Sensitivity (cut-off) if | ||||||||
| Specificity 95.0% | 233 | 85.0% (0.13)b (79.8–89.0) |
78.1% (0.25) (72.4–83.0) |
73.8% (1.13) (67.8–79.0) |
74.2% (16.4) (68.3–79.5) |
71.7% (0.42) (65.6–77.0) |
67.4% (7.25) (61.1–73.1) |
69.5% (0.47) (63.3–75.1) |
| Specificity 97.5% | 233 | 81.1% (0.17)b (76.6–85.7) |
75.1% (0.59)c (69.2–80.2) |
69.1% (0.77) (62.9–74.7) |
70.8% (24.0) (64.7–76.3) |
67.0% (1.23) (60.7–72.7) |
67.0% (8.11) (60.7–72.7) |
62.7% (0.89) (56.3–68.6) |
For all calculations, equivocal results were treated as positive. 95% CI, 95% confidence interval; LR+, positive likelihood ratio; ROC, receiver operator curve.
a
If defined by the manufacturer, the upper and lower limit of the equivocal zone are listed.
b
p < 0.05 vs all except Abbott.
c
p < 0.05 vs Euro S1.
d
p < 0.05 vs Diasorin.