Kidney Week

Abstract: FR-PO022

Correlation Between Incidence and Attributable Mortality Fraction of AKI: A Systematic Review

Session Information

• AKI: Epidemiology, Risk Factors, Prevention - II
  November 08, 2019 | Location: Exhibit Hall, Walter E. Washington Convention Center
  Abstract Time: 10:00 AM - 12:00 PM

Category: Acute Kidney Injury

• 101 AKI: Epidemiology, Risk Factors, and Prevention

Authors

• Komaru, Yohei, University of Tokyo, Tokyo, Japan

Yohei Komaru,

• Iwagami, Masao, University of Tsukuba, Tsukuba, Japan

Masao Iwagami,

• Hamasaki, Yoshifumi, University of Tokyo, Tokyo, Japan

Yoshifumi Hamasaki,

• Nangaku, Masaomi, University of Tokyo, Tokyo, Japan

Masaomi Nangaku,

• Doi, Kent, University of Tokyo, Tokyo, Japan

Kent Doi,

Background

The incidence and mortality of acute kidney injury (AKI) has extremely varied, even after the introduction of RIFLE, AKIN and KDIGO criteria. The quality of AKI diagnosis and management may also be influential. The present study aimed to investigate the association between AKI incidence and mortality of each cohort. We also investigated the effect of publication year and economic index on AKI mortality.

Methods

Our study aggregated the incidence and mortality of AKI through a systematic review of manuscripts on AKI patients diagnosed by Kidney Disease: Improving Global Outcomes (KDIGO)-equivalent criteria from 2004 to May, 2018. The search was conducted in MEDLINE, EMBASE, and Cochrane Library. We investigated the correlation between AKI incidence, mortality, and AKI attributable fraction of mortality. AKI attributable fraction of mortality in each cohort was calculated as follows: {(mortality of AKI patients)-(mortality of patients without AKI)} / (mortality of AKI patients). The impact of publication year and gross domestic product (GDP) on the mortality were also studied.

Results

The systematic review screened total 4149 manuscripts, and finally yielded 287 eligible cohorts (adults: 203 cohorts consisted of 7076459 patients; children: 84 cohorts of 69677 patients). In the adult cohorts, the mortality of AKI patients became higher (R²=0.023, β=0.12, P=0.033) but the attributable fraction of mortality otherwise decreased (R²=0.27, β=−0.43, P<0.001), as incidence of AKI augmented. Although the crude mortality of AKI patients decreased in more recent publications and in reports from higher GDP countries, the AKI attributable fraction did not decline in the same settings.

Conclusion

The AKI attributable fraction of mortality in the cohorts with high AKI incidence was relatively low, which possibly indicated the advantage of more experience in AKI diagnosis and management.