Table 4.
Characteristic | Univariate analysis | Full multivariable model† | Reduced multivariable model‡ | ||||||
---|---|---|---|---|---|---|---|---|---|
OR | 95% CI | P | Adjusted OR | 95% CI | P | Adjusted OR | 95% CI | P | |
Age ≥ 65 years | 7.6 | 4.5–13.1 | <0.001 | 2.7 | 0.6–12.3 | 0.21 | — | — | — |
Male | 1.2 | 0.8–2.0 | 0.36 | 1.3 | 0.4–4.7 | 0.70 | — | — | — |
COPD | 11.8 | 4.5–31.2 | <0.001 | 20.2 | 2.0–198.6 | 0.010 | 20.2 | 2.8–145.3 | 0.003 |
CCF | 12.0 | 5.2–27.8 | <0.001 | 12.1 | 1.5–99.8 | 0.021 | 7.8 | 1.2–49.2 | 0.028 |
Coronary artery disease | 5.2 | 2.4–11.4 | <0.001 | 7.8 | 0.7–82.7 | 0.09 | 12.9 | 1.7–98.9 | 0.014 |
Cerebrovascular disease | 7.5 | 2.1–26.5 | 0.002 | 3.6 | 0.1–108.3 | 0.46 | — | — | — |
CKD | 6.9 | 2.5–18.6 | <0.001 | 0.2 | 0.01–1.7 | 0.13 | — | — | — |
Diabetes mellitus | 3.6 | 1.9–6.8 | <0.001 | 8.8 | 1.7–45.9 | 0.010 | 7.4 | 1.6–33.9 | 0.010 |
Cancer | 2.3 | 1.1–4.7 | 0.021 | 0.6 | 0.1–3.4 | 0.58 | — | — | — |
Hypotension | 35.2 | 17.7–70.1 | <0.001 | 53.2 | 10.4–272.1 | <0.001 | 29.2 | 8.2–104.7 | <0.001 |
Hypoxia | 60.3 | 14.3–254.2 | <0.001 | 23.2 | 1.9–288.0 | 0.015 | — | — | — |
Bradycardia | 4.5 | 2.0–10.0 | <0.001 | 0.5 | 0.04–5.7 | 0.54 | — | — | — |
Platelets <180 × 109/L | 2.8 | 1.7–4.5 | <0.001 | 0.6 | 0.2–2.1 | 0.40 | — | — | — |
PT ≥ 20 s | 5.6 | 3.1–9.9 | <0.001 | 2.6 | 0.6–12.4 | 0.22 | — | — | — |
Albumin <34 g/L | 10.8 | 6.2–18.8 | <0.001 | 3.6 | 0.9–14.3 | 0.065 | 3.4 | 1.1–11.0 | 0.039 |
Creatinine ≥110 μmol/L | 17.9 | 9.9–32.8 | <0.001 | 1.0 | 0.3–3.9 | 0.96 | — | — | — |
Bicarbonate <22 mmol/L | 9.7 | 5.5–17.0 | <0.001 | 6.8 | 1.6–28.9 | 0.010 | 6.2 | 2.0–19.4 | 0.002 |
GGT < 135 IU/L | 5.3 | 3.2–8.8 | <0.001 | 4.5 | 1.0–19.2 | 0.044 | 5.1 | 1.5–17.6 | 0.010 |
Bilirubin <22 μmol/L | 3.1 | 1.9–5.1 | <0.001 | 3.9 | 0.9–17.5 | 0.075 | — | — | — |
All variables with P < 0.1 on univariate analysis, as well as prespecified additional variables of age ≥65 years and gender, were entered into a full multivariable logistic regression model.
The full multivariable model was entered into a backward stepwise logistic regression model with removal criterion P > 0.05, leaving a reduced model with all remaining variables with P < 0.05.
CCF, congestive cardiac failure; CI, confidence interval; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; GGT, gamma‐glutamyltransferase; OR, odds ratio; PT, prothrombin time.