Acute renal failure prolongs weaning from mechanical ventilation in critically ill patients

Acute renal failure (ARF) determines a worse prognosis in various medical scenarios. Since the syndrome of ARF can potentially interfere with the weaning from mechanical ventilation (MV), we sought to investigate whether the presence of ARF has any impact on weaning from MV. We studied 140 patients who received invasive MV for at least 48 hours in an oncologic ICU. Exclusion criteria: neurosurgical patients, pulmonary resections or strict end-of-life care. ARF definition: at least one value of serum creatinine (SCr) ≥ 1.5 mg/dl during the ICU stay. Patients were divided into ARF (n = 93) and non-ARF groups (NRF, n = 47). Criteria for weaning: PEEP ≤ 8 cmH₂O, pressure support ≤ 10 cmH₂O, FiO₂ ≤ 0.4 and spontaneous respiration. Primary endpoint: length of weaning. Secondary endpoints: length of MV, length of stay in ICU, and mortality. Groups were similar regarding age and gender. A higher number of ARF patients had hematological tumors (19.3 vs 6.4%, P = 0.04). The diagnosis of acute respiratory insufficiency (45 vs 44%) during the ICU stay and the diagnosis of ALI/ARDS as a cause for MV (18.2 vs 10.6%) did not differ between groups.

SAPS at entry was not different (48.1 ± 1.4 vs 43.5 ± 15.1) but ARF patients had markers of more severe disease in the long term: severe sepsis or septic shock (P < 0.0001); higher number of antibiotics (P = 0.0018); longer time of vasoactive drug (VAD) usage (P = 0.0005). Oliguria (urine output <500 ml/day, for at least 24 hours) was found in 47% of ARF patients, with the median time of 72 hours. SCr at ICU admission in the ARF group was 1.6 ± 0.1 mg/dl vs 0.7 ± 0.03 mg/dl in NRF. The total length of MV was higher in ARF patients (13 ± 11 vs 9 ± 6 days, P = 0.017). Moreover, ARF patients used higher levels of FiO₂ (highest FiO₂ 80, IQ 50–100 vs 65, IQ 50–100, P = 0.024). Less ARF patients reached criteria for weaning (42 vs 60%, P < 0.05), and had longer length for weaning from MV (83 ± 105 vs 33 ± 45 hours, P = 0.012). Cox regression analysis showed that an 85% increase in SCr (HR 2.30, CI 1.3–4.08), the presence of oliguria (HR 2.51, CI 1.24–5.08) and number of antibiotics greater than four (HR 2.64, CI 1.51–4.63) independently predicted weaning failure. The length of ICU stay (15 ± 12 vs 11 ± 7 days) and ICU mortality (67 vs 43%) were significantly higher in ARF patients. Multivariate analysis showed that oliguria (OR 27.3, 7.42–100.5), ARF (OR 2.43, 1.16–4.74) and prolonged use of VAD (OR 4.42, 1.28–15.2) were independent risk factors for mortality.

Increases in SCr, and particularly oliguria, during the ICU stay seriously impact on the duration of MV, weaning from MV and mortality in ICU patients. Although the presence of ARF appears to be a marker of a more severe condition, it is an independent negative factor for mechanically ventilated cancer patients.