Acute Kidney Injury As an Independent Predictor of Infection and Malignancy: the NARA-AKI Cohort Study

Background: Acute kidney injury (AKI) is associated with higher mortality and cardiovascular events. However, association between AKI and non-cardiac events such as infection or malignancy is largely unknown.

Methods: This is a retrospective cohort study. Inclusion criteria were adults who underwent non-cardiac surgery from 2007 to 2011 at Nara Medical University Hospital. Exclusion criteria were urological surgery, obstetric surgery, missing creatinine values peri-operatively, and pre-operative dialysis. The end of observation period was at the end of 2015 or loss to follow-up. A predictor was AKI defined by KDIGO criteria within 1-week post-operatively. Outcomes were hospitalization for infection or diagnoses of malignancy. Associations between AKI and outcomes were examined by Cox regression models.

Results: Among 6692 subjects, 445 (6.6%) developed AKI. During median follow-up of 4.0 years, there were 485 hospitalizations for infection and 1138 diagnoses of malignancy (2.0 and 5.1 events/100 patient-years, respectively). After adjustment for potential confounders, AKI was independently associated with hospitalization for infection and diagnoses of malignancy (Hazard ratio [95% confidence interval]: 1.64 [1.23-2.20] and 1.31 [1.06-1.61], respectively). Excluding recurrence of malignancy from outcomes and analyses limited to those who recover renal function by the time of discharge yielded similar results. Absolute lymphocyte counts were significantly lower and neutrophil-to-lymphocyte ratios were significantly higher among those with AKI.

Conclusions: AKI was significantly associated with hospitalization for infection and development of malignancy during long-term follow-up. Those with AKI might be in persistent immunosuppressed state.