Objectives: Acute kidney injury (AKI) worsens outcome in various scenarios. We sought to investigate whether the occurrence of AKI has any effect on weaning from mechanical ventilation.
Design and setting: Observational, retrospective study in a 23-bed medical/surgical intensive care unit (ICU) in a cancer hospital from January to December 2003.
Patients: The inclusion criterion was invasive mechanical ventilation for > or =48 hrs. AKI was defined as at least one measurement of serum creatinine of > or =1.5 mg/dL during the ICU stay. Patients were then separated into AKI and non-AKI patients (control group). The criterion for weaning was the combination of positive end-expiratory pressure of < or =8 cm H2O, pressure support of < or =10 cm H2O, and Fio2 of < or =0.4, with spontaneous breathing. The primary end point was duration of weaning and the secondary end points were rate of weaning failure, total length of mechanical ventilation, length of stay in the ICU, and ICU mortality.
Results: A total of 140 patients were studied: 93 with AKI and 47 controls. The groups were similar in regard to age, sex, and type of tumor. Diagnosis of acute lung injury/acute respiratory distress syndrome as cause of respiratory failure and Simplified Acute Physiology Score II at admission did not differ between groups. During ICU stay, AKI patients had markers of more severe disease: increased occurrence of severe sepsis or septic shock, higher number of antibiotics, and longer use of vasoactive drugs. The median (interquartile range) duration of mechanical ventilation (10 [6-17] vs. 7 [2-12] days, p = .017) and duration of weaning from mechanical ventilation (41 [16-97] vs. 21 [7-33.5] hrs, p = .018) were longer in AKI patients compared with control patients. Cox regression analysis demonstrated that a > or =85% increase in baseline serum creatinine (hazard rate, 2.30; 95% confidence interval, 1.30-4.08), oliguria (hazard rate, 2.51; 95% confidence interval, 1.24-5.08), and the number of antibiotics (hazard rate, 2.64; 95% confidence interval, 1.51-4.63) predicted longer duration of weaning. The length of ICU stay and ICU mortality rate were significantly greater in the AKI patients. After adjusting for Simplified Acute Physiology Score II, oliguria (odds ratio, 30.8; 95% confidence interval, 7.7-123.0) remained as a strong risk factor for mortality.
Conclusion: This study shows that renal dysfunction has serious consequences in the duration of mechanical ventilation, weaning from mechanical ventilation, and mortality in critically ill cancer patients.