Comparison between different equations for estimating glomerular filtration rate (GFR) with kinetic estimates of GFR in the critically ill

• Main Authors: Md Ralib, Azrina, Dzaharudin, Fatimah, Ishak, Nuraisyah, Mat Nor, Mohd Basri
• Format: Conference or Workshop Item
• Language: English
• Published: 2017
• Subjects: R Medicine (General)
• Online Access: http://irep.iium.edu.my/56693/26/56693_Comparison%20between%20different%20equations_new.pdf; http://irep.iium.edu.my/56693/

INTRODUCTION Accurate assessment of GFR in ICU patients is very important for institution of supportive therapy, preventive therapy early renal support, drug dosing modification or avoidance of nephrotoxic drugs and modification of drug dosing. Kinetic estimate of GFR (keGFR) takes into account the changes of creatinine over time, creatinine production rate, and the volume of distribution, hence postulated to be a more accurate estimate of GFR in the acute settings, where there are rapidly changing kidney functions as in the critically ill. We evaluated the association of the keGFR with estimated GFR by conventional method. METHODS This is an interim analysis of single centre, prospective observational study of critically ill patients. The study has been registered with the National Medical Research Register (NMRR-14-1897-21447) and has obtained ethics approval. Inclusion criteria were patients older than 18 years old with sepsis, defined as clinical infection and acute increase in SOFA score>2, and plasma procalcitonin>0.5ng/ml. Plasma creatinine were measured at seven time points, and eGFR were calculated by the Cockcroft-Gault, MDRD, CKD-EPI and keGFR equations. RESULTS Twenty-four patients were recruited so far, of which 10 (41.7%) had AKI. Two patients need dialysis, and one died. Ten patients developed AKI within 24 hours of ICU admission. keGFR strongly correlated with eGFR by CKD-EPI equation (r=0.92, p<0.0001) at all time points in all patients. Whereas, keGFR only moderately correlated with eGFR by MDRD and Cockcroft-Gault equations (r=0.79 and 0.77, respectively). eGFR by MDRD only strongly correlated with keGFR in subcohort of patients with AKI. CONCLUSIONS The new equation, keGFR strongly correlated with the eGFR by the CKD-EPI equation. In the absence of serial plasma creatinine measurement, eGFR can accurately be estimated by the CKD-EPI equation compared to others. Further analysis would involve analysis of the association with urinary creatinine clearance, and plasma Cystatin C.