Accuracy of clinical and radiographic variables as defined by Johanson et al11 for the diagnosis of VAP
| Variable or combination of variables | Sensitivity % (n) | Specificity % (n) | Positive predictive value % (n) | Negative predictive value % (n) |
|---|---|---|---|---|
| Chest radiograph (right or left) | 92 (12/13) | 33 (4/12) | 60 (12/20) | 80 (4/5) |
| Chest radiograph (right) | 73 (8/11) | 33 (4/12) | 50 (8/16) | 57 (4/7) |
| Chest radiograph (left) | 92 (11/12) | 75 (9/12) | 79 (11/14) | 90 (9/10) |
| Leucocytosis | 77 (10/13) | 58 (7/12) | 67 (10/15) | 70 (7/10) |
| Fever | 46 (6/13) | 42 (5/12) | 46 (6/13) | 42 (5/12) |
| Purulent secretions | 69 (9/13) | 42 (5/12) | 56 (9/16) | 56 (5/9) |
| Chest radiograph + one of three criteria | 85 (11/13) | 33 (4/12) | 58 (11/19) | 67 (4/6) |
| Chest radiograph + two of three criteria | 69 (9/13) | 75 (9/12) | 75 (9/12) | 69 (9/13) |
| Chest radiograph + all three criteria | 23 (3/13) | 92 (11/12) | 75 (3/4) | 52 (11/21) |
Sensitivity = rate of correctly classified patients with VAP and total number of patients with VAP; specificity = rate of correctly classified patients without VAP and total number of patients without VAP; positive predictive value = rate of correctly classified patients with VAP and total number of patients classified as having VAP (correctly positive plus false positive); negative predictive value = rate of correctly classified patients without VAP and total number of patients classified as not having VAP (correctly negative plus false negative).