Rationale: Understanding the changing incidence and impact of acute kidney injury requiring dialysis in patients with severe sepsis will allow better risk stratification, design of clinical trials, and guide resource allocation.
Objectives: To assess the longitudinal incidence of acute kidney injury requiring dialysis and its impact on mortality in patients with severe sepsis.
Methods: Retrospective cohort study of adults (≥20 yr) hospitalized with severe sepsis from 2000 to 2009 in the United States using a nationally representative database.
Measurements and main results: We calculated the incidences of acute kidney injury requiring dialysis and mortality over time. We used linear regression to assess temporal trends. We used logistic regression to estimate the odds of acute kidney injury requiring dialysis and mortality. Of the estimated 5,257,907 hospitalizations with severe sepsis, 6.1% had acute kidney injury requiring dialysis. The odds of acquiring acute kidney injury requiring dialysis increased by 14% in 2009 compared with 2000. Mortality in patients with acute kidney injury requiring dialysis was higher (43.6% vs. 24.9%; P < 0.001). After multivariable adjustment, odds of mortality declined 61% by the year 2009. Acute kidney injury requiring dialysis remained an independent predictor of mortality in patients with severe sepsis, although its influence on mortality declined with time.
Conclusions: Incidence of acute kidney injury requiring dialysis in patients with severe sepsis has increased over time; conversely, associated mortality has declined. The likelihood of demise from acute kidney injury requiring dialysis in patients with severe sepsis has also declined.
Keywords: acute kidney injury; dialysis; sepsis.