Ahead of Print
Impacto de Diferentes Fórmulas no Diagnóstico/Estadiamento de Doença Renal Crónica
Impact of Different Formulas on the Diagnosis/Staging of Chronic Kidney Disease
Main Article Content
Abstract
Introduction: Chronic kidney disease (CKD) affects approximately 9.8% of the Portuguese population. As it is a progressive condition and often asymptomatic in its early stages, screening and early diagnosis are essential. One of the main pillars is the estimation of the glomerular filtration rate (GFR), whose determination varies according to the equation used. We aimed to evaluate differences in CKD staging using validated equations for GFR calculation.
Methods: Observational, cross-sectional, retrospective, and analytical study conducted at USF Lethes (February 2025). Individuals aged ?18 years with a recorded serum creatinine value in SClínico® were included. Those without a recorded body weight were excluded. Sociodemographic, clinical, and anthropometric data were collected through the MIM@UF® platform and analyzed using LibreOffice®.
Results: Among 8 900 individuals, significant variations in CKD classification were observed depending on the equation used. The Cockcroft–Gault (CG) equation identified more cases (12%) than CKD-EPI 2009 (7.5%) and CKD-EPI 2021 (6.1%). CKD-EPI 2021 classified more individuals in the early stages (G1 and G2), whereas CG assigned a higher number to moderate stages (G3a and G3b). Greater agreement was found between CKD-EPI 2009 and 2021, particularly in transitions between stages.
Conclusion: The CG equation demonstrated higher sensitivity, while CKD-EPI 2021 showed greater specificity but a lower ability to detect early and moderate stages. These divergences in GFR estimation influence the diagnosis and staging of CKD, with a direct impact on clinical monitoring, treatment, and referral to Nephrology. Uniform adoption of CKD-EPI 2009, without a race factor, may improve diagnostic accuracy in primary health care in Portugal.
Article Details

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.