Case-control study: exact conditional logistic models regressing anti-PF4/heparin Ab status (dependent variable) on the periodontal status (grouped into tertiles)
| . | OR (95% CI) for anti-PF4/heparin Abs . | P . |
|---|---|---|
| Mean probing depth, mm | ||
| T1 (Ref.) | 1.00 | – |
| T2 | 2.12 (0.56; 10.52) | .28 |
| T3 | 7.12 (1.73; 46.13) | .005 |
| Extent probing depth ≥ 3 mm, % | ||
| T1 (Ref.) | 1.00 | – |
| T2 | 4.50 (1.28; 22.59) | .02 |
| T3 | 3.68 (1.02; 18.52) | .05 |
| Extent probing depth ≥ 4 mm, % | ||
| T1 (Ref.) | 1.00 | – |
| T2 | 14.74 (2.77; 370.44) | < .001 |
| T3 | 14.30 (2.75; 354.53) | < .001 |
| Extent probing depth ≥ 5 mm, % | ||
| T1 (Ref.) | 1.00 | – |
| T2 | 5.54 (1.40; 32.91) | .01 |
| T3 | 6.32 (1.87; 31.82) | .002 |
| . | OR (95% CI) for anti-PF4/heparin Abs . | P . |
|---|---|---|
| Mean probing depth, mm | ||
| T1 (Ref.) | 1.00 | – |
| T2 | 2.12 (0.56; 10.52) | .28 |
| T3 | 7.12 (1.73; 46.13) | .005 |
| Extent probing depth ≥ 3 mm, % | ||
| T1 (Ref.) | 1.00 | – |
| T2 | 4.50 (1.28; 22.59) | .02 |
| T3 | 3.68 (1.02; 18.52) | .05 |
| Extent probing depth ≥ 4 mm, % | ||
| T1 (Ref.) | 1.00 | – |
| T2 | 14.74 (2.77; 370.44) | < .001 |
| T3 | 14.30 (2.75; 354.53) | < .001 |
| Extent probing depth ≥ 5 mm, % | ||
| T1 (Ref.) | 1.00 | – |
| T2 | 5.54 (1.40; 32.91) | .01 |
| T3 | 6.32 (1.87; 31.82) | .002 |
ORs for the presence of anti-PF4/heparin Abs with their 95% CI (median-unbiased, mid-P-method) and P values are given; n = 40 matched pairs. Models were adjusted for age (continuous), sex, school education, and smoking status. Tertiles for mean probing depth were: ≤ 2.24 mm, 2.25-2.73 mm, and 2.75-4.75 mm; for extent probing depth ≥ 3 mm: ≤ 27.3%, 27.5%-52.3%, and 55.5%-89.3%; for extent probing depth ≥ 4 mm: 0%-2.3%, 2.5%-22.9%, and 23.1%-68.8%; for extent probing depth ≥ 5 mm: 0%, 1.8%-5.6%, and 5.8%-53.8%.
OR indicates odds ratio; CI, confidence interval; T, tertile; and –, not applicable.