Table 2.
Univariate risk factors of a prolonged length of ICU stay.
| Variable outcomes | AUC for ROC | 95% CI | Crude OR | 95% CI | Sensitivity (%) | Specificity (%) |
|---|---|---|---|---|---|---|
| Age (years) | ||||||
| <60 | 1.0 | |||||
| ≥60 | 0.629 | 0.543–0.716 | 2.881** | 1.518–5.467 | 62.1 | 63.7 |
| NYHA classification | ||||||
| I-II | 1.0 | |||||
| III-IV | 0.589 | 0.499–0.678 | 2.496* | 1.231–5.063 | 36.4 | 81.4 |
| Ejection fraction (%) | ||||||
| >40 | 1.0 | |||||
| ≤40 | 0.562 | 0.472–0.653 | 2.400* | 1.052–5.474 | 42.4 | 79.4 |
| Hematocrit (%) | ||||||
| >33 | 1.0 | |||||
| ≤33 | 0.570 | 0.483–0.657 | 3.355* | 1.202–9.366 | 90.9 | 19.6 |
| PaO2/FiO2 ratio (mmHg) | ||||||
| >200 | 1.0 | |||||
| ≤200 | 0.607 | 0.517–0.696 | 2.838** | 1.420–5.674 | 40.9 | 80.4 |
| Platelet count (/mm3) | ||||||
| >120,000 | 1.0 | |||||
| ≤120,000 | 0.607 | 0.517–0.696 | 2.838** | 1.420–5.674 | 40.9 | 80.4 |
| Inotropic drugs (items) | ||||||
| 1 | 1.0 | |||||
| ≥2 | 0.672 | 0.586–0.758 | 4.932*** | 2.463–9.876 | 53.0 | 81.4 |
| Creatinine (mg%) | ||||||
| <1.2 | 1.0 | |||||
| ≥1.2 | 0.581 | 0.491–0.671 | 2.508* | 1.192–5.277 | 31.8 | 84.3 |
| Potassium level (mEq/L) | ||||||
| >3.2 | 1.0 | |||||
| ≤3.2 | 0.578 | 0.487–0.668 | 2.812* | 1.251–6.323 | 27.3 | 88.2 |
| Atrial fibrillation grading# | ||||||
| <2 | 1.0 | |||||
| ≥2 | 0.634 | 0.547–0.721 | 3.020** | 1.588–5.742 | 59.1 | 67.6 |
ICU: intensive care unit; AUC for ROC: area under receiver operating characteristic curve; CI: confidence interval; OR: odds ratio; NYHA: New York Heart Association; PaO2: partial pressure of arterial oxygen; FiO2: fraction of inspired oxygen in a gas mixture.
#Atrial fibrillation (AF) grading: 0: no AF; 1: AF without tachyarrhythmia and needs no treatment; 2: AF with tachyarrhythmia and needs antiarrhythmic therapy; 3: AF with tachyarrhythmia and hemodynamic instability and needs antiarrhythmic therapy.
*P < 0.05, **P < 0.01, ***P < 0.001.