KDIGO stage is the maximum of two independently evaluated axes, never a sum.

SERUM-CREATININE AXIS: 1.5–1.9× baseline, or a rise of ≥ 0.3 mg/dL, is Stage 1; 2.0–2.9× baseline is Stage 2; ≥ 3.0× baseline, renal replacement therapy initiated, an estimated GFR < 35 mL/min/1.73 m² (patients under 18 years only), or a creatinine ≥ 4.0 mg/dL is Stage 3. The ≥ 4.0 mg/dL route is not standalone: KDIGO requires the AKI definition (a rise of ≥ 0.3 mg/dL, or ≥ 1.5× baseline) to be met first, so a chronically elevated creatinine that never rose is not Stage 3 AKI. With no baseline entered, ≥ 4.0 mg/dL reports Stage 3 flagged as not settled.
URINE-OUTPUT AXIS: the four Table 2 rows are (rate, duration) pairs, not rate bands, and each is tested independently with the highest satisfied row governing. < 0.5 mL/kg/h for 6 to under 12 hours is Stage 1; < 0.5 mL/kg/h for 12 hours or more is Stage 2; < 0.3 mL/kg/h for 24 hours or more is Stage 3; anuria for 12 hours or more is Stage 3. Branching on the rate first is the classic defect: 0.25 mL/kg/h for 8 hours is Stage 1, not Stage 3. A rate < 0.5 mL/kg/h held for less than 6 hours meets no row. Anuria is a clinical flag, because KDIGO defines no millilitre figure and none is invented here, and it necessarily satisfies the < 0.5 mL/kg/h rows too, so anuria for 6 to under 12 hours is Stage 1.
UNRESOLVABLE AXIS: when the duration is missing, or “12 hours or more” is chosen with a rate < 0.3 mL/kg/h leaving the 24-hour row open, the stage shown is the highest certain stage with a “≥” flag, which is KDIGO’s own Table 10 notation. No duration is ever guessed, and the flag is set only where an open row could change the answer.

Stage 0 means the KDIGO definition is not met on the criteria entered, which is not proof that AKI is absent. Higher stage associates with mortality and renal replacement therapy in the outcome literature, but the staging is a classification, not a treatment threshold. BASELINE CREATININE IS THE HARDEST INPUT. With no prior value, enter the LOWEST creatinine of this admission: in 710 critically ill children with true baselines it detected AKI with sensitivity 87.8% and specificity 71.0% (Lee 2022, 7-day window). Lee 2022 is DOI 10.23876/j.krcp.21.120. Do not back-calculate from an assumed GFR of 75, which is KDIGO’s own appendix suggestion: in the same children it was 31.5% sensitive, missed roughly two thirds of the AKI, and put AKI incidence at 19.1% against a true 58.7%. The direction reverses between adults and children, so adult reassurance that back-calculation over-diagnoses must not be carried across, and under-staging is the dangerous direction in a PICU. Record which window the entered value came from. KDIGO defines the ≥ 0.3 mg/dL rise as a rise WITHIN 48 HOURS; this calculator applies it as current minus the baseline you enter, with no window, so a baseline from weeks or months back can produce Stage 1 from a rise that was never acute. There is no paediatric modification of the urine-output thresholds. pRIFLE is a separate instrument and is neither reproduced nor blended in here. The bedside Schwartz equation (0.413 × height in cm ÷ serum creatinine in mg/dL) behind the eGFR branch was validated at roughly 1 to 16 years, so do not extrapolate it to neonates. KDIGO does not state which weight indexes the mL/kg/h [NEEDS SOURCE], nor is there a KDIGO-endorsed paediatric baseline rule [NEEDS SOURCE]. Conversion uses 1 mg/dL = 88.42 µmol/L with two-decimal rounding so KDIGO’s printed SI equivalents stage as intended (353.6 µmol/L resolves to 4.00 mg/dL). The urine-output duration is entered as the Table 2 bands rather than free hours because an hours box invites false precision about a figure read off a nursing chart.

Result	Interpretation
< 1	Stage 0 (no AKI by KDIGO criteria) — The KDIGO 2012 definition of acute kidney injury is not met on the criteria entered. Interpret in the full clinical context; absence of a criterion here reflects the data provided, not proof that AKI is absent. Check whether the result is flagged as a lower bound: when a low urine output was entered without a duration window, the urine-output axis could not be evaluated at all, and that is not the same finding as no AKI.
1 to <2	Stage 1 — KDIGO Stage 1 — the least severe AKI category: serum creatinine 1.5–1.9× baseline or a rise of ≥ 0.3 mg/dL, or urine output < 0.5 mL/kg/h sustained for 6 hours to under 12 hours. Higher stages are associated with worse outcomes in the literature; the stage is a descriptive classification, not a treatment threshold.
2 to <3	Stage 2 — KDIGO Stage 2 — an intermediate AKI category: serum creatinine 2.0–2.9× baseline, or urine output < 0.5 mL/kg/h sustained for 12 hours or more.
≥ 3	Stage 3 — KDIGO Stage 3 — the most severe AKI category: serum creatinine ≥ 3.0× baseline, or ≥ 4.0 mg/dL once the AKI definition itself is met, initiation of renal replacement therapy, urine output < 0.3 mL/kg/h for 24 hours or more, anuria for 12 hours or more, or — in a patient under 18 years — an estimated GFR < 35 mL/min/1.73 m². Check whether the result is flagged as not settled: a creatinine of 4.0 mg/dL or above entered with no baseline reaches this stage on the value alone, and a baseline showing no acute rise would take it back out of AKI altogether.

References

KDIGO Acute Kidney Injury Work Group. KDIGO Clinical Practice Guideline for Acute Kidney Injury. Kidney Int Suppl. 2012;2(1):1–138. Definition = Rec 2.1.1; staging = Rec 2.1.2 / Table 2 (p. 19); indeterminate-staging precedent = Chapter 2.4, Table 10 (p. 30).Primary source of record for every staging threshold, the max-of-two-axes rule, the four (rate, duration) urine-output rows, and the '≥ 1' / '?' notation used when an axis cannot be resolved.DOI 10.1038/kisup.2012.1
Palevsky PM, et al. Reading between the (guide)lines — the KDIGO practice guideline on acute kidney injury in the individual patient. Kidney Int. 2014;85(1):49–61.Reproduces KDIGO Table 2 including the '<18 years, eGFR < 35' Stage-3 branch. Corroborating secondary source ONLY. Its urine-output rows are laid out as a rate ladder, which is an easy thing to implement from by mistake; the (rate, duration) row structure this calculator implements is taken from the primary guideline itself (first reference above), not from this reproduction.Source
Schwartz GJ, Muñoz A, Schneider MF, et al. New equations to estimate GFR in children with CKD. J Am Soc Nephrol. 2009;20(3):629–637.Bedside equation eGFR = 0.413 × height(cm) ÷ SCr(mg/dL) used by the Stage-3 pediatric branch; validated ~1–16 y.PMID 19158356 DOI 10.1681/ASN.2008030287
Palevsky PM, et al. KDOQI US Commentary on the 2012 KDIGO Clinical Practice Guideline for Acute Kidney Injury. Am J Kidney Dis. 2013;61(5):649–672.National-society commentary confirming the KDIGO definition and staging.PMID 23499048 DOI 10.1053/j.ajkd.2013.02.349
Lee YJ, Park YS, Park SJ, Jhang WK. Comparison of methods for estimating baseline serum creatinine to predict acute kidney injury in critically ill children. Kidney Res Clin Pract. 2022;41(3):322–331.PRIMARY support for the surrogate-baseline guidance, and it is PAEDIATRIC — 710 patients aged 1 month to 18 years, single centre, all with a measured baseline within 3 months to compare against. The lowest creatinine within 7 days of PICU admission performed best (ICC 0.62; AKI sensitivity 87.8%, specificity 71.0%; misclassification 19.2%; kappa 0.60; incidence 63.5% against a true 58.7%, a slight OVER-estimate). Back-calculation from an assumed eGFR was far worse and worse in the dangerous direction (sensitivity 31.5%, specificity 98.3%, misclassification 40.3%; incidence 19.1% against the same true 58.7%). The paper contrasts this with adult reports of back-calculation OVER-estimating AKI — the direction reverses in children. Note the 7-day window is the paper's choice, not a standard.DOI 10.23876/j.krcp.21.120
Cooper DJ, Plewes K, Grigg MJ, Patel A, Rajahram GS, William T, Hiemstra TF, Wang Z, Barber BE, Anstey NM. An Evaluation of Commonly Used Surrogate Baseline Creatinine Values to Classify AKI During Acute Infection. Kidney Int Rep. 2021;6(3):645–656.SECONDARY SUPPORT — the paediatric Lee 2022 above is the primary support here, and this is kept only for what Lee does not test. It compared MDRD against CKD-EPI, an assumed GFR of 100 as well as KDIGO's suggested 75, and age/sex-standardised reference tables: every method built on an assumed GFR of 75 missed over half of all AKI; CKD-EPI at an assumed GFR of 100 tracked overall incidence best but still misassigned stages; the lowest creatinine measured during the admission over-called AKI by about a fifth yet correlated best with the reference value. CAUTION — 247 ADULTS with Plasmodium knowlesi malaria in Malaysian Borneo, so adult single-infection evidence, no longer relied on for any paediatric claim. It is also the reason the notes do not present 'the adult literature' as uniform: this adult cohort found back-calculation UNDER-detecting AKI, the same direction Lee found in children, not the over-estimation Lee contrasts against.PMID 33732979 DOI 10.1016/j.ekir.2020.12.020

Reproduction rights

Freely reproducible.KDIGO AKI staging is a set of factual numeric cut-points and mathematical rules (multipliers, absolute SCr/eGFR/UO thresholds, durations); facts and mathematical criteria are not copyrightable and may be implemented directly with attribution. No proprietary response-descriptor prose is reproduced — every option label and explanation here is written in this project's own words. The bedside Schwartz equation is likewise a formula (kdigo-aki.md IP status).

Independent clinical validation: pending

Two independent clinical validators will be named here once review is complete.

Validator slot open 1

Validator slot open 2

2026-08-10v1.0.0Initial releaseInitial published text.
2026-09-03v1.1.0Formula correctionReads a birthday entered in DAYS as that birthday. Age is stored in years and days convert at 365.25, so a sixth birthday of 2191 days became 5.9986 years and a tenth of 3652 days became 9.9986. Anything that floors or bands on whole years then read the child as a year younger. The conversion now snaps to a whole year when the day count is within one day of one, which is the largest the drift can be: 365.25 averages the leap cycle exactly, so a true birthday is always within 0.75 days of the integer. Ages entered in years or months were already exact and are unchanged, and an age more than a day from a birthday is untouched. Found 2026-09-03 by an independent recompute of every calculator from its published source. On this score an eighteenth birthday entered in days left KDIGO's under-18 estimated-GFR route to Stage 3 open on an adult.
2026-09-03v1.2.0Formula correctionStages a recorded urine output of zero exactly as it stages the anuria box. A five-year-old with 0 mL/kg/h sustained 12 hours or more returned Stage 2 flagged as a floor with the box unticked, and Stage 3 with it ticked — the same patient one stage apart on whether a second control was used, in what this score's own notes call the dangerous direction. Table 2's fourth row is anuria for 12 hours or more, and a rate of zero is anuria: there is no urine. The score already entailed anuria into the rows built on 0.5 mL/kg/h so that a patient described in words was not under-staged against one described in millilitres; this closes the other half. NO THRESHOLD IS INVENTED, which is the reason anuria remains a separate clinical flag: zero is not a cutoff KDIGO declined to give. Any positive rate is unchanged and still needs the box — 0.01 mL/kg/h is oliguria, not anuria. Founder decision of 2026-09-03, from an independent recompute of every calculator against its published source.
2026-09-06v1.2.1ClarificationAdded a one-line description for the catalogue card and shortened field guidance to fit an info toggle. No rule, threshold or reference changed.

Other renal and metabolic scores

Important. For use by qualified health professionals as an informational and educational aid. It supports clinical judgment and does not replace it. Verify every result independently before making a clinical decision. This is not a medical device.