Kidney Week

Abstract: FR-PO0400

Should We Eliminate KDIGO Stage 2 from AKI Classification?

Session Information

• AKI: Biomarkers, Diagnostics, and Risk Prediction
  October 23, 2026 | Location: Exhibit Hall A, Convention Center
  Abstract Time: 10:00 AM - 12:00 PM

Category: Acute Kidney Injury

• 102 AKI: Clinical, Outcomes, and Trials

Authors

• Acosta-Ochoa, Isabel, Hospital Rio Carrion, Palencia, CL, Spain

Isabel Acosta-Ochoa, MD, PhD

• Hernández García, Esther, Hospital Rio Carrion, Palencia, CL, Spain

Esther Hernández García, MD

• Sanchez Gil, Jimmy Reinaldo, Hospital General de Fuerteventura, Puerto del Rosario, CN, Spain

Jimmy Reinaldo Sanchez Gil, MD

• Rodriguez, Astrid, Hospital Clinico Universitario de Valladolid, Valladolid, CL, Spain

Astrid Rodriguez

• Oviedo, Victoria, Hospital Rio Carrion, Palencia, CL, Spain

Victoria Oviedo, MD

• Chandu Nanwani, Avinash, Hospital General de Fuerteventura, Puerto del Rosario, CN, Spain

Avinash Chandu Nanwani

• Falconi Sarmiento, Angelo R., Hospital Rio Carrion, Palencia, CL, Spain

Angelo R. Falconi Sarmiento, MD

• Gonzalez, Carmen Calderon, Hospital Rio Carrion, Palencia, CL, Spain

Carmen Calderon Gonzalez, DrMed

• Ampuero, Karina, Hospital Rio Carrion, Palencia, CL, Spain

Karina Ampuero, MD

• Velasco, Rubén, Hospital Rio Carrion, Palencia, CL, Spain

Rubén Velasco, MD

• Coca, Armando, Hospital Clinico Universitario de Valladolid, Valladolid, CL, Spain

Armando Coca, MD, MPH

Background

KDIGO AKI guidelines include an AKI severity classification which is the most used worldwide (an update is expected at the end of 2026). Stages are classified into 3 categories, Stage 1 (S1) and 3 (S3) have sub-criteria, Stage 2 (S2) only a SCr increment criterion. Some authors propose that S2 should be merged into S3; they argue that S2 normally has a smaller number of patients and they exhibit similar mortality and hard outcomes compared to S3. We tested this theory in a real-world cohort.

Methods

Retrospective, observational, multicentric study (3 centers, one in island territory) of in-patients with AKI diagnosed by KDIGO guidelines, treated by nephrology, during a 4y period. We compared outcomes: time to nephrology consultation (TtNC), length of stay (LOS) and mortality between S1 and S2 and S2 and S3 groups. We excluded initiation of RRT (because it is a S3 criterion). We further analyzed individuals that met S2 by rate of SCr increment between 2.0-2.9x that were re-classified into S3.

Results

We included 2896 AKI episodes that occurred in 1828 individuals. 1253 (43%) S1, 430 (15%) S2, 1213 (42%) S3. Comparing S2 to S3, we found no significant differences in age, HTN, DM, CAD prevalence and ICU admission. In S3 male sex, Charlson Index, CKD status and hospitalization in medical wards were significantly higher, while hospital acquired AKI was significantly lower. In outcomes we observed no differences in TtNC, in S3 LOS was longer [17±15 vs. 19±16; p=0,03] and mortality was higher [62 (14%) vs. 307 (25%); p<0,001]. When comparing S1 to S2 we observed no differences in TtNC, S2 had longer LOS [13±12 vs.17±15; p<0,001] and higher mortality [91 (7%) vs. 62 (14%); p<0,001].
When testing only SCr increment as the AKI S2 criterion, 716 (25%) cases would have been catalogued in this stratum; we found that 286 patients were reclassified as S3 because 271 reached a SCr >4.0mg and/or 91 initiated RRT.

Conclusion

The rate of individuals classified in S2 is small but not insignificant. In our cohort 40% of patients with a SCr increment between 2.0-2.9x were reclassified into S3 because they initiated RRT and/or reached a SCr >4.0 mg. Patients classified as S2 show shorter LOS and less mortality than S3 patients, but longer LOS and higher mortality, when compared to S1. With these findings we conclude that S2 adds granularity in AKI studies and evaluation; Stage 2 shouldn't be erased or merged.