Table 2. Multivariable analysis for prediction of 28-day mortality in patients with acute melioidosis.
| Variable | Adjusted Odds Ratio (95% CI) | P value | |
|---|---|---|---|
| Age | 1.0 (0.99–1.1) | 0.29 | |
| Diabetes | 0.56 (0.24–1.3) | 0.2 | |
| Pre-existing renal disease | 3.9 (1.5–10.5) | 0.007 | |
| Bp cell response | 0.56 (0.36–0.88) | 0.01 | |
| Neutrophil count / μl | > 4000–8000 | 1 | 1 |
| ≤ 4000 | 5.8 (1.1–29.6) | 0.04 | |
| > 8000–12,000 | 8.1 (2.1–31.5) | 0.002 | |
| > 12,000–20,000 | 7.5 (1.8–30.6) | 0.005 | |
| > 20,000 | 18.9 (3.1–114.2) | 0.001 | |
The table displays the odds ratios for variables studied as predictors of mortality in 200 patients with acute melioidosis in a multivariable logistic regression model. Bp cell response is the log10 transformed ELIspot response in fresh peripheral blood mononuclear cells to heat-killed B. pseudomallei. Neutrophil count represents the peripheral blood neutrophil count with > 4000 to 8000 neutrophils / μl as the comparator group. Odds ratios are adjusted for the other variables included in the model.