Table III.
Performance of allergy tests in the diagnosis of peanut allergy (N = 104)
| Single diagnostic test | Correct diagnoses∗ | No. of false positives | No. of false negatives | No. of BATs | No. of OFCs | Change in the no. of OFC† |
|---|---|---|---|---|---|---|
| SPT | 78 (75) | 1 (1) | 1 (1) | — | 24 (23) | −12 (−33) |
| P-sIgE | 57 (55) | 3 (3) | 3 (3) | — | 41 (39) | +5 (+13) |
| Ara h 2 | 82 (79) | 1 (1) | 2 (2) | — | 19 (18) | −17 (−46) |
| BAT‡ | 89 (86§) | 2 (2) | 1 (1) | 104 (100) | 12 (12) | −24 (−67) |
| Combination of diagnostic tests | Correct diagnoses∗ | No. of false positives | No. of false negatives | No. of BATs | No. of OFCs | Change in the no. of OFCs† |
|---|---|---|---|---|---|---|
| SPT + P-sIgE | 67 (64) | 1 (1) | 0 (0) | — | 36 (35) | 0 (0) |
| P-sIgE + Ara h 2 | 66 (63) | 1 (1) | 1 (1) | — | 36 (35) | 0 (0) |
| P-sIgE + BAT | 66 (63) | 1 (1) | 2 (2) | 104 (100) | 35 (34) | −1 (−3) |
| SPT + BAT | 77 (74) | 2 (2) | 0 (0) | 104 (100) | 25 (25) | −11 (−31) |
| SPT + Ara h 2 | 78 (75) | 1 (1) | 0 (0) | — | 25 (24) | −11 (−31) |
| Ara h 2 + BAT | 77 (74) | 1 (1) | 2 (2) | 104 (100) | 24 (24) | −12 (−33) |
| SPT + P-sIgE + Ara h 2 | 67 (64) | 1 (1) | 0 (0) | — | 36 (35) | 0 (0) |
| SPT + Ara h 2 + BAT | 70 (67) | 1 (1) | 0 (0) | 104 (100) | 33 (33) | −3 (−8) |
| SPT + P-sIgE + BAT | 63 (61) | 2 (2) | 0 (0) | 104 (100) | 39 (38) | +3 (+8) |
| P-sIgE + Ara h 2 + BAT | 63 (61) | 1 (1) | 1 (1) | 104 (100) | 39 (38) | +3 (+8) |
| SPT + P-sIgE + Ara h 2 + BAT | 60 (58) | 1 (1) | 0 (0) | 104 (100) | 43 (42) | +7 (+19) |
| BAT as a second step in the diagnostic process | Correct diagnoses∗ | No. of false positives | No. of false negatives | No. of BATs | No. of OFCs | Change in the no. of OFCs† |
|---|---|---|---|---|---|---|
| SPT → BAT | 98 (94) | 3 (3) | 2 (2) | 24 (23) | 1 (1) | −35 (−97) |
| P-sIgE → BAT | 93 (89) | 5 (5) | 3 (3) | 41 (39) | 3 (3) | −33 (−92) |
| Ara h 2 → BAT | 99 (95) | 2 (2) | 2 (2) | 19 (18) | 1 (1) | −35 (−97) |
| SPT + P-sIgE → BAT | 97 (93) | 3 (3) | 1 (1) | 36 (35) | 3 (3) | −33 (−92) |
| SPT + Ara h 2 → BAT | 99 (95) | 2 (2) | 1 (1) | 38 (37) | 2 (2) | −34 (−94) |
| SPT + P-sIgE + Ara h 2 → BAT | 96 (92) | 3 (3) | 1 (1) | 36 (35) | 4 (4) | −32 (−89) |
| BAT as a third step in the diagnostic process | Correct diagnoses∗ | No. of false positives | No. of false negatives | No. of BATs | No. of OFCs | Change in the no. of OFCs† |
|---|---|---|---|---|---|---|
| SPT → Ara h 2 → BAT | 98 (94) | 3 (3) | 3 (3) | 6 (6) | 0 (0) | −36 (−100) |
Results are presented as number of patients (% of total study population).
Ara h 2, sIgE to Ara h 2.
∗
The proportion of correct diagnoses was determined as (“true-positives” + “true-negatives”)/104.
†
Reduction in the number of OFCs was calculated in comparison with the number of OFCs after SPT and sIgE (ie, 36 OFCs, row in italic); negative numbers represent a decrease and positive numbers an increase in the number of OFCs required.
‡
For BAT, we used 4.78% for the average of CD63+ basophils at 10 and 100 ng/mL of peanut extract as the diagnostic cutoff point.
§
For BAT, excluding nonresponders, the proportion of correct diagnoses is 96.7%.