Analysis of the prospective validation cohort (January 2017 to November 2019; n = 687): CLIA HIT-IgG testing with positive results according to Figure 2 (n = 45 of 687 [6.6%])
| CLIA HIT-IgG . | ID-PaGIA-H/PF4 . | 4T score . | Posttest probability . | Clinical prediction . | HIPA . | Conclusion (n) . | |||
|---|---|---|---|---|---|---|---|---|---|
| U/mL . | n . | Titer . | n . | Pretest probability . | n . | P (95% CI) . | HIT . | ||
| Routine evaluation according toFigure 2 | |||||||||
| >3.0 | 45 | Neg | 1 | Low | 1 | 0% (0-2)* | Excluded | Negative | Accurately excluded (1)† |
| 4 | 1 | Low | 1 | 35% (17-59)* | Possible | Positive | Solved by HIPA (1) | ||
| Not done‡ | 43 | Intermediate | 34 | 100% (79-100) | Predicted | 32 positive, 2 negative§ | Accurately predicted (32), probably accurate prediction (2) | ||
| High | 9 | 100% (98-100) | Predicted | Positive | Accurately predicted (9) | ||||
| Post hoc evaluation of the 43 samples with CLIA HIT-IgG > 3.0 U/mL and intermediate to high 4T score‡ | |||||||||
| 43 | 2 | 1 | Intermediate | 1 | 76% (30-96)* | Predicted | Negative§ | Probably accurate prediction (1) | |
| 4 | 4 | Intermediate | 4 | 98% (94-99)* | Predicted | 3 positive, 1 negative§ | Accurately predicted (3), probably accurate prediction (1) | ||
| 8 | 8 | Intermediate | 7 | 100% (99-100)* | Predicted | Positive | Accurately predicted (7) | ||
| High | 1 | 100% (100-100)* | Predicted | Positive | Accurately predicted (1) | ||||
| 16 | 7 | Intermediate | 6 | 100% (100-100)* | Predicted | Positive | Accurately predicted (6) | ||
| High | 1 | 100% (100-100)* | Predicted | Positive | Accurately predicted (1) | ||||
| >16 | 23 | Intermediate | 16 | 100% (100-100)* | Predicted | Positive | Accurately predicted (16) | ||
| High | 7 | 100% (100-100)* | Predicted | Positive | Accurately predicted (7) | ||||
| CLIA HIT-IgG . | ID-PaGIA-H/PF4 . | 4T score . | Posttest probability . | Clinical prediction . | HIPA . | Conclusion (n) . | |||
|---|---|---|---|---|---|---|---|---|---|
| U/mL . | n . | Titer . | n . | Pretest probability . | n . | P (95% CI) . | HIT . | ||
| Routine evaluation according toFigure 2 | |||||||||
| >3.0 | 45 | Neg | 1 | Low | 1 | 0% (0-2)* | Excluded | Negative | Accurately excluded (1)† |
| 4 | 1 | Low | 1 | 35% (17-59)* | Possible | Positive | Solved by HIPA (1) | ||
| Not done‡ | 43 | Intermediate | 34 | 100% (79-100) | Predicted | 32 positive, 2 negative§ | Accurately predicted (32), probably accurate prediction (2) | ||
| High | 9 | 100% (98-100) | Predicted | Positive | Accurately predicted (9) | ||||
| Post hoc evaluation of the 43 samples with CLIA HIT-IgG > 3.0 U/mL and intermediate to high 4T score‡ | |||||||||
| 43 | 2 | 1 | Intermediate | 1 | 76% (30-96)* | Predicted | Negative§ | Probably accurate prediction (1) | |
| 4 | 4 | Intermediate | 4 | 98% (94-99)* | Predicted | 3 positive, 1 negative§ | Accurately predicted (3), probably accurate prediction (1) | ||
| 8 | 8 | Intermediate | 7 | 100% (99-100)* | Predicted | Positive | Accurately predicted (7) | ||
| High | 1 | 100% (100-100)* | Predicted | Positive | Accurately predicted (1) | ||||
| 16 | 7 | Intermediate | 6 | 100% (100-100)* | Predicted | Positive | Accurately predicted (6) | ||
| High | 1 | 100% (100-100)* | Predicted | Positive | Accurately predicted (1) | ||||
| >16 | 23 | Intermediate | 16 | 100% (100-100)* | Predicted | Positive | Accurately predicted (16) | ||
| High | 7 | 100% (100-100)* | Predicted | Positive | Accurately predicted (7) | ||||
Calculated as with CLIA HIT-IgG value 1.0 to 3.0 U/mL (LR, 63.13) (Table 2).
Sample with a high titer of antiphospholipid antibodies leading to a false-positive CLIA HIT-IgG (3.99 U/mL; PV-17.199).
Routine diagnostic evaluation without PaGIA according to Figure 2.
Might be false-negative HIPA testing: positive ELISA inhibited by high-dose heparin, clinical course suggestive of HIT (PV-19.014 and PV-19.232; supplemental Table 5).