The role of peritoneal dialysis in the treatment of acute kidney injury in neurocritical patients: a retrospective Brazilian study

Abstract

Background: Acute kidney injury (AKI) occurs frequently in the neurocritical intensive care unit and is associated with greater morbidity and mortality. AKI and its treatment, including acute kidney replacement therapy, can expose patients to a secondary greater brain injury. This study aimed to explore the role of peritoneal dialysis (PD) in neurocritical AKI patients in relation to metabolic and fluid control, complications related to PD and outcome. Methods: Neurocritical AKI patients were treated by PD (prescribed Kt/V = 0.40/session) using a flexible catheter and a cycler and lactate as a buffer. Results: A total of 58 patients were included. The mean age was 61.8 ± 13.2 years, 65.5% were in the intensive care unit, 68.5% needed intravenous inotropic agents, 72.4% were on mechanical ventilation, APACHE II was 16 ± 6.67 and the main neurological diagnoses were stroke (25.9%) and intracerebral haemorrhage (31%). Ischaemic acute tubular necrosis (iATN) was the most common cause of AKI (51.7%), followed by nephrotoxic ATN AKI (25.8%). The main dialysis indications were uraemia and hypervolemia. Blood urea and creatinine levels stabilised after four sessions at around 48 ± 11 mg/dL and 2.9 ± 0.4 mg/dL, respectively. Negative fluid balance and ultrafiltration increased progressively and stabilised around 2.1 ± 0.4 L /day. Weekly delivered Kt/V was 2.6 ± 0.31. The median number of high-volume PD sessions was 6 (4–10). Peritonitis and mechanical complications were not frequent (8.6% and 10.3%, respectively). Mortality rate was 58.6%. Logistic regression identified as factors associated with death in neurocritical AKI patients: age (odds ratio (OR) = 1.14, 95% confidence interval (CI) = 1.09–2.16, p = 0.001), nephrotoxic AKI (OR = 0.78, 95% CI = 0.69– 0.95, p = 0.03), mechanical ventilation (OR = 1.54, 95% CI = 1.17–2.46, p = 0.01), intracerebral haemorrhage as main neurological diagnoses (OR = 1.15, 95% CI = 1.09–2.11, p = 0.03) and negative fluid balance after two PD sessions (OR = 0.94, 95% CI = 0.74–0.97, p = 0.009). Conclusion: Our study suggests that careful prescription may contribute to providing adequate treatment for most neurocritical AKI patients without contraindications for PD use, allowing adequate metabolic and fluid control, with no increase in the number of infectious, mechanical and metabolic complications. Mechanical ventilation, positive fluid balance and intracerebral haemorrhage were factors associated with mortality, while patients with nephrotoxic AKI had lower odds of mortality compared to those with septic and ischaemic AKI. Further studies are needed to investigate better the role of PD in neurocritical patients with AKI.