Abstract
Background and Aims
Studies have shown that millions of hospitalized patients suffer from Acute Kidney Injury (AKI) per year which increases mortality risk for these patients. Different definitions for AKI have been proposed during the past years such as RIFLE (2002) and AKIN (2004). In 2012, KDIGO published a clinical practice guideline harmonizing AKIN and RIFLE into one general guideline which classifies AKI into 3 stages, where stage 1 is defined as an absolute increase of SCr ≥ 0.3 mg/dl over 48 hours or a relative increase in SCr ≥ 50% from baseline within the previous 7 days. A recent study [Sparrow et al., 2019] evaluated the impact of further categorizing AKI stage 1 into 2 stages based on SCr criteria. The study separates KDIGO AKI stage 1 and AKIN stage 1 into 2 stages (KDIGO-4 and AKIN-4) based on the different SCr criteria. Having different AKI definitions makes it challenging to analyze AKI incidence and associated outcomes among studies. The present study aimed to investigate the incidence of AKI events defined by 4 different definitions (standard AKIN and KDIGO, and modified AKIN-4 and KDIGO-4) and its association with in-hospital mortality.
Method
Retrospective clinical data available for all adult (≥18 years old) hospital admissions to a local health district in Athens, Greece between October 1999 and March 2019 was used in the analysis. We excluded patients whose time between admission and discharge was less than 7 days. Also, patients with less than 5 Scr measurements were omitted from the analysis resulting in the final cohort of 7242 admissions. We used the AKIN, KDIGO, AKIN-4, and KDIGO-4 definitions to check the incidence of AKI. As our second goal, we assessed associations of AKI-events with in-hospital mortality, adjusted for characteristics (age, sex, AKI staging) using multivariable logistic regression.
Results
The incidence of in-hospital AKI using the modified KDIGO-4 was 6.72% for stage 1a, 15.71% for stage 1b, 8.06% for stage 2, and 2.97% for stage 3; however, these percentages for AKIN-4 were 11.5%, 5.83%,1.75%, and 0.33% for stage 1a, stage 1b, stage 2, and stage 3, respectively. Using the standard KDIGO and AKIN definition, 19.08 and 14.05 % developed stage 1, respectively. To find the association between AKI stages and in-hospital mortality, we considered the most severe stage of AKI reached by a patient. Results of logistic regression models show that in-hospital mortality increased as the stage of AKI events increased for both KDIGO-4 and AKIN-4 (Table 1). Table 2 shows the same results using the original KDIGO and AKIN definitions.
Conclusion
The results of both definitions (AKIN-4 and KDIGO-4) show a significant association with mortality, but KDIGO-4 has a larger odds ratio meaning that AKI classification based on KDIGO-4 has a stronger association with mortality than AKI classification based on AKIN-4. However, based on our results, splitting stage 1 to stage 1a and stage 1b does not seem to make a difference; hence, using KDIGO-4 as a replacement for KDIGO would not have a significant impact on capturing AKI events.
The impact of local health professions education grants: is it worth the investment?
