The real importance of pre-existing comorbidities on long-term mortality after acute kidney injury

PLoS One. 2012;7(10):e47746. doi: 10.1371/journal.pone.0047746. Epub 2012 Oct 17.


Background: The causes of death on long-term mortality after acute kidney injury (AKI) have not been well studied. The purpose of the study was to evaluate the role of comorbidities and the causes of death on the long-term mortality after AKI.

Methodology/principal findings: We retrospectively studied 507 patients who experienced AKI in 2005-2006 and were discharged free from dialysis. In June 2008 (median: 21 months after AKI), we found that 193 (38%) patients had died. This mortality is much higher than the mortality of the population of São Paulo City, even after adjustment for age. A multiple survival analysis was performed using Cox proportional hazards regression model and showed that death was associated with Khan's index indicating high risk [adjusted hazard ratio 2.54 (1.38-4.66)], chronic liver disease [1.93 (1.15-3.22)], admission to non-surgical ward [1.85 (1.30-2.61)] and a second AKI episode during the same hospitalization [1.74 (1.12-2.71)]. The AKI severity evaluated either by the worst stage reached during AKI (P=0.20) or by the need for dialysis (P=0.12) was not associated with death. The causes of death were identified by a death certificate in 85% of the non-survivors. Among those who died from circulatory system diseases (the main cause of death), 59% had already suffered from hypertension, 34% from diabetes, 47% from heart failure, 38% from coronary disease, and 66% had a glomerular filtration rate <60 previous to the AKI episode. Among those who died from neoplasms, 79% already had the disease previously.

Conclusions: Among AKI survivors who were discharged free from dialysis the increased long-term mortality was associated with their pre-existing chronic conditions and not with the severity of the AKI episode. These findings suggest that these survivors should have a medical follow-up after hospital discharge and that all efforts should be made to control their comorbidities.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Acute Kidney Injury / mortality*
  • Aged
  • Brazil / epidemiology
  • Cause of Death
  • Chronic Disease
  • Comorbidity*
  • Female
  • Hospitalization / statistics & numerical data
  • Humans
  • Kaplan-Meier Estimate
  • Liver Diseases / mortality
  • Male
  • Proportional Hazards Models
  • Time Factors

Grant support

Mariana B Pereira received a Grant from CAPES (“Coordenação de Aperfeiçoamento de Pessoal de Nível Superior”). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.