Table 1.
Data are Mean ± 1SD for continuous variables and number of patients (% of group total) positive for each category
| Characteristic | Clinical setting | P-valuea | Mortality as in-patient in ICU only | ||||
|---|---|---|---|---|---|---|---|
| CS n = 151 n (%) |
ICU n = 150 n (%) |
Unadjusted OR (95% CI) P-value |
Adjusted OR (95% CI) P-value |
||||
| Age, yr | 68 ± 10 | 55 ± 17 | <.001 | 1.04 (1.01–1.07) | 0.002 | 1.03 (1.00–1.06) | <.001 |
| Male | 110 (73) | 96 (64) | 0.14 | 1.38 (0.60–3.17) | 0.44 | – | – |
| Smoking (ever) | 80 (53) | 63 (42) | 0.057 | 0.62 (0.27–1.40) | 0.24 | – | – |
| CKD | 38 (25) | 8 (5) | <.001 | 1.19 (0.23–6.17) | 0.83 | – | – |
| Diabetes | 38 (25) | 34 (23) | 0.03 | 2.0 (0.87–4.79) | 0.10 | 1.48 (0.58–3.74) | 0.22 |
| CCF | 36 (24) | 5 (3) | <.001 | 2.45 (0.39–15.3) | 0.33 | – | – |
| Hypertension | 88 (58) | 35 (23) | <.001 | 1.31 (0.54–3.16) | 0.55 | – | – |
| Sepsis | 2 (1) | 41 (27) | <.001 | 0.99 (0.41–2.36) | 0.99 | – | – |
| Diuretic use | 60 (39) | 3 (2) | <.001 | – | – | – | – |
| AKI diagnosis based on | |||||||
| SCr alone | 36 (24) | 48 (32) | 0.11 | – | – | – | – |
| Urine output cons | 46 (31) | 55 (37) | 0.25 | – | – | – | – |
| Urine output mean | 104 (69) | 72 (48) | <.001 | – | – | – | – |
| bWald, F-pr for AKI (Model 1) | 66.9, 2df < .001 | 15.1, <.001 | – | – | – | – | – |
| bWald, F-pr for AKI (Model 2) | 67.3, 2df < .001 | – | – | – | – | – | – |
aStatistical differences between groups of patients on admission to either cardiac surgery (CS) or intensive care unit (ICU) were assessed by Students t-test (age only) or chi-squared test for categorical data. bAssessment of the proportion of patients diagnosed as having AKI based on differing clinical criteria (SCr, UOcons, UOmean) were assessed by logistic regression fitting the outcome (AKI, yes/no) with a binomial error distribution. Wald statistic and F-probability are given after correction for any significant confounders (e.g. Model 1: age & diabetes, Model 2: age, diabetes & diuretic use). Model 2 not conducted for ICU patients as so few were on diuretics. Statistical significance was accepted at P < 0.05. SCr, serum creatinine; UO, urine output; urine output cons, output determined by volume produced in consecutive hours; urine output mean, output determined by average volume per hour. All data analyses were conducted using Genstat v17 (VSNi, UK). For mortality data in ICU, referent categories were coded as 0 = No-Diabetes or No-AKI