Table 4.
Effect of early surgery on 30-day or 90-day mortality: logistic regression analysis.
| Categories of mortality | Method of adjustment | odds ratio | 95% confidence interval | p-value | |
|---|---|---|---|---|---|
| 30-day mortality | Non-adjusted | None | 0.707 | (0.218, 2.287) | 0.563 |
| Adjusted | Model 1 | 0.394 | (0.083, 1.867) | 0.241 | |
| Model 2 | 0.002 | (<0.001, 0.347) | 0.018 | ||
| 90-day mortality | Non-adjusted | None | 0.741 | (0.298, 1.846) | 0.520 |
| Adjusted | Model 3 | 0.588 | (0.180, 1.922) | 0.379 | |
| Model 4 | 0.194 | (0.026, 1.436) | 0.108 | ||
Model 1: adjusted for age, Child-Turcotte-Pugh, and bacteremia.
Model 2: adjusted for age, Child-Turcotte-Pugh class, Child-Turcotte-Pugh score, MELD score, hepatocellular carcinoma, GI bleeding, prothrombin time (INR), sepsis, urinary tract infection, other musculoskeletal infection, and multiple spinal lesion.
Model 3: adjusted for MELD score, presence of combined infection, and multiple spinal lesion.
Model 4: adjusted for age, Child-Turcotte-Pugh class, Child-Turcotte-Pugh score, MELD score, hepatocellular carcinoma, GI bleeding, platelet, bilirubin, prothrombin time (INR), creatinine, sepsis, presence of combined infection, urinary tract infection, other musculoskeletal infection, and multiple spinal lesion.