#1900 Should we prescind of KDIGO Stage 2 in AKI severity classification?

Abstract

Abstract Background and Aims The most worldwide used classification system for staging AKI is KDIGO-2012, it was based on a combination of the RIFLE and AKIN criteria. KDIGO severity stages are defined: Stage 1 ↑ in SCr by >0.3 mg/dl or increase in SCr to >1.5x baseline. Stage 2 ↑ in SCr ≥2.0–2.9x baseline. Stage 3 ↑ in SCr ≥3.0x baseline, ↑ in SCr to >4.0 mg/dl, or initiation of RRT. Some authors advocate for the elimination of the Stage 2 level and its incorporation in the severer stage; the justification is that this stage has a smaller number of individuals, and they show poorer clinical outcomes. We tested this hypothesis in a real-world setting. Method We retrospectively included all patients admitted to nephrology ward and nephrology consultations with AKI diagnosis by KDIGO-2012 classification, during a 3-y period. We compared clinical outcomes between KDIGO-AKI Stage 2 and 3. We excluded need for HD (because it is a criterion for Stage 3) and HD dependence at discharge (a consequence of initiating HD). In addition, we calculated the proportion of patients classified in Stage 2 by the rate of SCr increment, that were re-classified into Stage 3 by KDIGO criteria. Results 1130 individuals met inclusion criteria. 458 (41%) Stage-1, 148 (13%) Stage-2, 524 (46%) in Stage-3. We found no statistically significant differences in age, Charlson's Index, HTN and DM prevalence, hospitalization in medical wards, and hospital acquired AKI between groups. We found significant differences in sex distribution (61% Stage-2, 70% Stage-3) (see Table A: Features). Hospital stay was significantly shorter in Stage1-15d (±12) but not significantly different between Stage-2 19d (±16) and Stage-3 20d (±17). Median of time to nephrology consultation was 6-7d in all groups. Rate of in-hospital mortality incremented gradually in every severity stage: 9% Stage-1, 18% Stage-2 and 31% Stage-3 (see Table B: Results). When using only the rate of SCr increment as the classification criterion, 265 (24%) of patients would have been catalogued in the moderate AKI stratum. Of the 117 patients that were reclassified as AKI Stage-3, 36 needed HD, and 81 reached a SCr >4.0 mg/dl. Conclusion The proportion of subjects classified in KDIGO AKI Stage-2 is small but not negligible. We found that Stage-2 individuals show longer hospital stay and increased mortality than Stage-1, and slightly better outcomes than Stage-3; therefore, we consider that an increase in SCr ≥2.0–2.9x from baseline still leads to deleterious consequences. We also observed that possibly KDIGO AKI Stage 3 criteria lead to a skew to classification in the severer stage, (almost half of individuals are reclassified). With these findings, should we prescind of KDIGO Stage-2 in AKI severity classification?