Diagnostic accuracy of different ELISA assays by manufacturer
| Assay . | No. studies . | No. patients . | Statistical model . | Sensitivity (95% CI) . | Specificity (95% CI) . | Positive LR (95% CI) . | Negative LR (95% CI) . |
|---|---|---|---|---|---|---|---|
| Polyspecific ELISA, low threshold | |||||||
| Pooled estimates | 31* | 8933* | Unified | 96.7 (89.7-99.0) | 86.8 (82.0-90.5) | 7.3 (5.4-10.0) | 0.04 (0.01-0.12) |
| GTI19,20,28,41,46,47,50,54,61,68,74,75,79 † | 13 | 3804 | Unified | 99.9 (90.9-100.0) | 87.4 (79.2-92.7) | 7.9 (4.7-13.4) | 0.001 (<0.001-0.11) |
| Zymutest63,69,77 | 3 | 295 | Fixed effects‡ | 99.5 (94.8-100.0) | 77.6 (72.9-82.3) | 4.1 (3.1-5.5) | 0.05 (0.01-0.12) |
| Asserachrom39,45,46,49-51,53,59,60,63,64,71,75,78,79 | 15 | 2621 | Unified | 92.7 (73.6-98.3) | 87.3 (79.9-92.3) | 7.3 (4.6-11.7) | 0.08 (0.02-0.34) |
| In-house assay16,43,48,52,79 § | 5 | 2673 | Unified | 93.2 (76.3-98.3) | 92.6 (90.1-94.0) | 12.6 (9.7-16.3) | 0.07 (0.02-0.28) |
| Polyspecific ELISA, intermediate threshold | |||||||
| Pooled estimates* | 5* | 2334* | Unified | 98.6 (95.2-99.6) | 94.8 (92.2-96.5) | 18.8 (12.7-27.9) | 0.01 (0.004-0.05) |
| GTI19,47,61,65 | 5 | 2646 | Unified | 97.4 (92.4-99.1) | 95.9 (94.3-97.0) | 23.6 (17.2-32.4) | 0.03 (0.009-0.08) |
| Zymutest77 | 1 | 87 | — | 100.0 | 82.5 | 5.7 | 0.00 |
| Asserachrom65 | 1 | 1291 | — | 99.6 | 93.6 | 15.6 | 0.01 |
| IgG-specific ELISA, low threshold | |||||||
| Pooled estimates* | 12* | 3116* | Unified | 98.3 (95.1-99.4) | 85.4 (78.2-90.6) | 6.7 (4.5-10.2) | 0.02 (0.01-0.05) |
| GTI55,63,64,70 | 4 | 639 | Unified | 99.6 (22.7-100.0) | 89.9 (86.2-92.6) | 9.8 (7.2-13.4) | 0.004 (<0.001-3.7) |
| Zymutest44,63,69,70,77 | 5 | 829 | Unified | 99.2 (86.4-100.0) | 85.8 (77.1-91.5) | 7.0 (4.3-11.4) | 0.008 (<0.001-0.18) |
| Asserachrom66,71 | 2 | 160 | Fixed effects‡ | 72.0 (68.4-75.5) | 93.8 (90.3-97.4) | 3.6 (2.04-5.6) | 0.27 (0.10-0.72) |
| Technozym72 | 1 | 41 | — | 100.0 | 0.92 | 12.7 | 0.00 |
| In-house assay19,20 § | 4 | 2958 | Unified | 99.2 (96.4-99.8) | 83.5 (71.5-91.0) | 6.0 (3.4-10.7) | 0.009 (0.002-0.04) |
| IgG-specific ELISA, intermediate threshold | |||||||
| Pooled estimates* | 4* | 2545* | Unified | 91.2 (86.3-94.5) | 93.5 (89.1-96.2) | 14.1 (8.1-24.5) | 0.09 (0.06-0.15) |
| Zymutest77 | 1 | 87 | — | 100.0 | 82.5 | 5.7 | 0.00 |
| In-house assay19,20 § | 5 | 4416 | Unified | 89.8 (82.7-94.2) | 96.0 (94.5-97.1) | 22.5 (16.3-31.2) | 0.10 (0.06-0.18) |
| Assay . | No. studies . | No. patients . | Statistical model . | Sensitivity (95% CI) . | Specificity (95% CI) . | Positive LR (95% CI) . | Negative LR (95% CI) . |
|---|---|---|---|---|---|---|---|
| Polyspecific ELISA, low threshold | |||||||
| Pooled estimates | 31* | 8933* | Unified | 96.7 (89.7-99.0) | 86.8 (82.0-90.5) | 7.3 (5.4-10.0) | 0.04 (0.01-0.12) |
| GTI19,20,28,41,46,47,50,54,61,68,74,75,79 † | 13 | 3804 | Unified | 99.9 (90.9-100.0) | 87.4 (79.2-92.7) | 7.9 (4.7-13.4) | 0.001 (<0.001-0.11) |
| Zymutest63,69,77 | 3 | 295 | Fixed effects‡ | 99.5 (94.8-100.0) | 77.6 (72.9-82.3) | 4.1 (3.1-5.5) | 0.05 (0.01-0.12) |
| Asserachrom39,45,46,49-51,53,59,60,63,64,71,75,78,79 | 15 | 2621 | Unified | 92.7 (73.6-98.3) | 87.3 (79.9-92.3) | 7.3 (4.6-11.7) | 0.08 (0.02-0.34) |
| In-house assay16,43,48,52,79 § | 5 | 2673 | Unified | 93.2 (76.3-98.3) | 92.6 (90.1-94.0) | 12.6 (9.7-16.3) | 0.07 (0.02-0.28) |
| Polyspecific ELISA, intermediate threshold | |||||||
| Pooled estimates* | 5* | 2334* | Unified | 98.6 (95.2-99.6) | 94.8 (92.2-96.5) | 18.8 (12.7-27.9) | 0.01 (0.004-0.05) |
| GTI19,47,61,65 | 5 | 2646 | Unified | 97.4 (92.4-99.1) | 95.9 (94.3-97.0) | 23.6 (17.2-32.4) | 0.03 (0.009-0.08) |
| Zymutest77 | 1 | 87 | — | 100.0 | 82.5 | 5.7 | 0.00 |
| Asserachrom65 | 1 | 1291 | — | 99.6 | 93.6 | 15.6 | 0.01 |
| IgG-specific ELISA, low threshold | |||||||
| Pooled estimates* | 12* | 3116* | Unified | 98.3 (95.1-99.4) | 85.4 (78.2-90.6) | 6.7 (4.5-10.2) | 0.02 (0.01-0.05) |
| GTI55,63,64,70 | 4 | 639 | Unified | 99.6 (22.7-100.0) | 89.9 (86.2-92.6) | 9.8 (7.2-13.4) | 0.004 (<0.001-3.7) |
| Zymutest44,63,69,70,77 | 5 | 829 | Unified | 99.2 (86.4-100.0) | 85.8 (77.1-91.5) | 7.0 (4.3-11.4) | 0.008 (<0.001-0.18) |
| Asserachrom66,71 | 2 | 160 | Fixed effects‡ | 72.0 (68.4-75.5) | 93.8 (90.3-97.4) | 3.6 (2.04-5.6) | 0.27 (0.10-0.72) |
| Technozym72 | 1 | 41 | — | 100.0 | 0.92 | 12.7 | 0.00 |
| In-house assay19,20 § | 4 | 2958 | Unified | 99.2 (96.4-99.8) | 83.5 (71.5-91.0) | 6.0 (3.4-10.7) | 0.009 (0.002-0.04) |
| IgG-specific ELISA, intermediate threshold | |||||||
| Pooled estimates* | 4* | 2545* | Unified | 91.2 (86.3-94.5) | 93.5 (89.1-96.2) | 14.1 (8.1-24.5) | 0.09 (0.06-0.15) |
| Zymutest77 | 1 | 87 | — | 100.0 | 82.5 | 5.7 | 0.00 |
| In-house assay19,20 § | 5 | 4416 | Unified | 89.8 (82.7-94.2) | 96.0 (94.5-97.1) | 22.5 (16.3-31.2) | 0.10 (0.06-0.18) |
Studies conducted in the same study population were included only once to maintain independency of observations.
Including technically identical Gen-Probe assay.
In tests with <4 studies available, summary estimates were calculated using a fixed-effects model (Mantel-Haenszel method; to be interpreted with caution).
Application of pooled “in-house assay” data to individual tests must be done with caution because these tests are constructed differently (every assay must be evaluated individually).