Table 4
Predictors of mortality by univariate and multivariable logistic regression analysis
| Variables | Mortality | Univariate | Multivariable | |||
|---|---|---|---|---|---|---|
| Yes (n = 188) | No (n = 248) | OR (CI 95%) | p | OR (CI 95%) | p | |
Male gender n/total (%) | 56/188 (29.8%) | 56/248 (22.6%) | 1.45 (0.94–2.24) | 0.89 | ||
Age, years (mean ± SD) | 93.9 ± 3.4 | 93.3 ± 3.4 | 1.05 (0.99–1.11) | 0.96 | 1.12 (1.04–1.21) | 0.002 |
Charlson’s score ≥ 6 n/total (%) | 144/188 (76.6%) | 111/248 (44.8%) | 1.90 (1.30–2.80) | 0.01 | 1.69 (1.01–2.83) | 0.049 |
Length of hospital stay, days (mean ± SD) | 21.2 ± 20.3 | 13.1 ± 11.2 | 1.04 (1.02–1.05) | < 0.001 | ||
ICU admission n/total (%) | 170/188 (90.4%) | 162/248 (65.3%) | 5.01 (2.89–8.70) | < 0.001 | ||
Mechanical ventilation n/total (%) | 87/188 (46.3%) | 9/248 (3.6%) | 22.87 (11.08–47.21) | < 0.001 | 5.29 (2.24–12.46) | < 0.001 |
Vasopressors n/total (%) | 116/187 (62.0%) | 21/248 (8.5%) | 17.66 (10.33–30.17) | < 0.001 | 6.82 (3.55–13.11) | < 0.001 |
| AKI all cases | 131/188 (69.7%) | 65/248 (26.2%) | 6.47 (4.25–9.85) | < 0.001 | ||
| AKI KDIGO 1 | 49/188 (26.1%) | 47/248 (18.9%) | 3.35 (2.03–5.51) | < 0.001 | 2.01 (1.09–3.72) | 0.025 |
| AKI KDIGO 2 | 38/188 (20.2%) | 11/248 (4.4%) | 11.09 (5.32–23.11) | < 0.001 | 4.95 (2.05–11.94) | < 0.001 |
| AKI KDIGO 3 | 44/188 (23.4%) | 7/248 (2.8%) | 20.18 (8.61–47.27) | < 0.001 | 6.52 (2.37–17.98) | < 0.001 |
| Community AKI | 14/188 (7.4%) | 31/248 (12.5%) | 1.45 (0.72–2.91) | 0.297 | ||
| Hospital AKI | 97/188 (51.6%) | 28/248 (11.3%) | 11.12 (6.65–18.61) | < 0.001 | ||
| Community and Hospital AKI | 20/188 (10.6%) | 6/248 (2.4%) | 10.70 (4.10–27.94) | < 0.001 | ||
| RRT | 13/188 (6.9%) | 0/248 (0.0%) | (−) | < 0,001 | ||
Legend: RRT Renal Replacement Therapy, OR Odds ratio, CI Confidence interval, SD Standard deviation, ICU Intensive care unit, AKI Acute kidney injury, KDIGO Kidney Disease Improving Global Outcomes. Note: the following variables entered the multivariable logistic regression model: age, Charlson’s score ≥ 6, length of hospital stay, ICU admission, mechanical ventilation, vasopresssors, AKI stage and RRT. The last column only shows the ones that remained statistically significant at the last step of the multivariable model
Bold entries are variables with p values < 0.10 on univariate analyses were selected for inclusion in stepwise backward multivariable logistic regression models and variables with p values < 0.05 in final analyses were considered statistically significant