CKD-EPI 2021 eGFR
2021 CKD-EPI creatinine equation without a race coefficient, indexed to 1.73 m² body surface area.
- Version
- 0.1
- Updated
- July 17, 2026
- Automated tests
- Included
- Sources
- 2
Evidence ledger · Public by default
This ledger keeps source selection, calculation or decision-rule logic, automated checks, intended population, limitations, and version history visible in one place.
2021 CKD-EPI creatinine equation without a race coefficient, indexed to 1.73 m² body surface area.
Additive 0–9 point factor score: CHF 1, hypertension 1, age 65–74 1 or ≥75 2, diabetes 1, stroke/TIA/thromboembolism 2, vascular disease 1, female sex category 1.
Additive 0–9 point bleeding-risk factor score with renal and liver dysfunction, drug exposure, and alcohol counted separately.
Seven-item Wells PE score with the two-tier threshold: PE unlikely ≤4; PE likely >4.
One point each for confusion, urea >7 mmol/L, respiratory rate ≥30/min, low blood pressure, and age ≥65.
One point each for respiratory rate ≥22/min, systolic blood pressure ≤100 mm Hg, and altered mentation.
Five 0–2 point domains: history, ECG, age, atherosclerotic risk factors, and troponin relative to the assay reference limit; total range 0–10.
Eight binary criteria: age ≥50, pulse ≥100/min, oxygen saturation <95%, unilateral leg swelling, hemoptysis, recent surgery/trauma, prior VTE, and exogenous hormone use.
Best eye response (1–4) + verbal response (1–5) + motor response (1–6); total range 3–15 with component notation retained.
Six organ-system subscores (respiratory, coagulation, hepatic, cardiovascular, neurologic, renal), each 0–4; total range 0–24.
Five 1–3 point domains: bilirubin, albumin, PT prolongation/INR, ascites, and encephalopathy; class A 5–6, B 7–9, C 10–15.
Cockcroft–Gault (1976), with actual, Devine ideal, and adjusted body weight 0.4 shown separately.
Measured sodium + correction factor × ((glucose − 100) / 100), comparing Katz 1.6 and Hillier 2.4 mEq/L per 100 mg/dL.
Mosteller, Du Bois & Du Bois, and Haycock equations displayed side by side.
Devine ideal body weight, adjusted body weight using a 0.4 correction factor, and BMI displayed separately.
Common simplified Payne-derived adjustment: total Ca + 0.8 × (4 − albumin g/dL), or SI equivalent +0.02 × (40 − albumin g/L).
Bazett, Fridericia, Framingham, and Hodges corrections displayed together; RR is derived as 60/heart rate.
Na − (Cl + HCO3), optional K inclusion, and optional albumin correction AG + 2.5 × (4 − albumin g/dL).
2Na + glucose/18 + BUN/2.8, optional measured gap, and optional ethanol contributions using divisors 4.6 and 3.7.
ANC = WBC (cells/µL) × (segmented neutrophils % + bands %) / 100.
k = ln(C1/C2)/(t2−t1); half-life = ln(2)/k; optional first-order forward projection from the second level.
Two-level first-order elimination plus steady-state intermittent-infusion equations for Cmax, Cmin, Vd, CL, AUCτ, AUC24, and accumulation.
Steady-state one-compartment intermittent-infusion model using two post-distribution levels; AUC24 = (dose/CL) × (24/interval).
Steady-state two-level Sawchuk–Zaske-style one-compartment workup with optional user-entered regimen simulation using patient-specific k and Vd.
Steady-state two-level Sawchuk–Zaske-style one-compartment workup with optional user-entered amikacin regimen simulation using patient-specific k and Vd.
Matzke population k = 0.00083 × CrCl + 0.0044, editable Vd coefficient × weight, and finite-infusion first-dose and steady-state one-compartment projections.
Drug-specific high-dose protocol: gentamicin/tobramycin 7 mg/kg or amikacin 15 mg/kg using actual weight unless actual weight exceeds 120% of IBW, then AdjBW = IBW + 0.4(ABW−IBW); initial interval selected from protocol CrCl bands.
One random concentration 6–14 hours after infusion start is plotted against a linear digitization of the published Hartford figure. Gentamicin/tobramycin require 7 mg/kg; the separately labeled amikacin adaptation uses 15 mg/kg and plots measured concentration ÷ 2. A ±0.2 mg/L line margin selects the longer interval.
Searchable evidence synthesis mapping grouped drugs to warm, cold, either, or unspecified compresses and to agent-specific antidotes or interventions. No antidote preparation or dosing is generated.
Reaction-phenotype triage plus a simplified R1 side-chain relationship map. Cross-reactivity estimates are shown only with the allergy-verification and structural-similarity populations from which they were derived.
Searchable synthesis separating simulated Y-site physical compatibility from manufacturer-supported use of LR as an infusion diluent, with formulation, concentration, contact-time, and age exceptions shown in the result row.
Publication standard
Original equation, current guidance, validation studies, intended population, exclusions, units, and known controversies are recorded.
Reference cases and boundary conditions are checked against the cited equations or decision rules.
Wording, applicability, limitations, units, and high-risk misuse cases are kept beside the tool.
The evidence version, calculation version, update date, and material changes stay linked.
Restricted release · Evidence remains visible
Generic albumin adjustment can misclassify calcium status, so the calculator resets to a locked state on every visit, prioritizes ionized calcium when accuracy matters, and never labels its output as a measurement.
RxTray publishes source-backed clinical reference software with documented tests, versions, and limitations. No interface or evidence record should be treated as a substitute for professional judgment, current labeling, or institutional policy.