Differences among six prevalent creatinine clearance calculation methods by covariate modeling of CL for Netilmicin using NONMEM for inference
Oteo I, Jauregizar N, Astobieta A, Rodriguez Sasiain JM, Lukas C.J, Calvo R
University of the Basque Country, DynaKin.S.L
Objectives: To evaluate the ability of 6 different methods in estimating creatinine clearance (CrCL) as predictor of clearance (CL) within a pharmacokinetic (PK) model for netilmicin, a potentially nephrotoxic aminoglycoside drug. .
Methods: Plasma levels (n= 310) and serum creatinine levels, from 62 adult patients treated with netilmicin (a single dose of 100 mg) for short-term prophylaxis after minor urological surgery as previously analyzed in Jauregizar et al (2003)2 were used. Patients were in good general health, and had normal renal function (serum creatinine ≤ 1.9 mg/dL). Here, CrCL was estimated by Cockroft & Gault (1976)3; Mawer et al. (1972)4; MDRD formulae5; DAF method (2010)6; Jelliffe et al. (1973)1 in addition to the method used in Jaureguizar et al (2003)2 (Mawer method corrected by ideal weigth, calculated from Peck's formula7 (MM)). The NONMEM® objective function (OFV) was then used to compare the predictive ability (for the CL of Netilmicin in this population) of the 6 methods of CrCL calculation using the same model (CL=qslope*CrCL). The OFV is approximately chi-square distributed and a change of 3.8 is significant at p<0.05. A third order GAM was also applied in covariate fits for all methods with CrCL as the sole variable.
Results: Preliminary ANOVA analysis showed that CrCL estimation using MM method resulted statistically different when compared with DAF and MDRD methods. In the same way, Jelliffe was different compared with Mawer, DAF and MDRD methods. The MDRD had OFV= 173.342 (vs. 141.858 for MM) with other methods lying in between but also significantly less predictive than MM. Similarly, the interindividual variability was higher for MDRD compared with MM 38.60% vs 29.22% respectively). The parameter qslope varied from 0.051 - 0.066 among methods, this fact could affect variably to netilmicin clearance.
Conclusions: The use of alternative methods for CrCL calculation for a specific drug and patient population is not justified without model specific development.
References:
[1] Marx et al. Evaluation of the Cockroft-Gault, Jelliffe and Wright formulae in estimating renal function in elderly cancer patients. Ann Oncol. 2004 Feb;15(2):291-5.
[2] Jauregizar et al. Population pharmacokinetics of netilmicin in short-term prophylactic treatment . Br J Clin Pharmacol. 2003 Jun;55(6):552-9.
[3] Spruill et al. Comparison of estimated glomerular filtration rate with estimated creatinine clearance in the dosing of drugs requiring adjustments in elderly patients with declining renal function. Am J Geriatr Pharmacother. 2008 Aug;6(3):153-60.
[4] Mawer et al. Computer-assisted prescribing of kanamycin for patients with renal insufficiency. Lancet. 1972 Jan 1;1(7740):12-5
[5] Li et al. Diagnostic accuracy of various glomerular filtration rates estimating equations in patients with chronic kidney disease and diabetes. Chin Med J (Engl). 2010 Mar 20;123(6):745-51
[6] Diamandopoulos et al. Comparison of estimated creatinine clearance among five formulae (Cockroft-Gault, Jelliffe, Sanaka, simplified 4-variable MDRD and DAF) and the 24hours-urine-collection creatinine clearance. Hippokratia. 2010 Apr; 14(2):98-104.
[7] ABBOTTBASE Pharmacokinetic Systems. Operations Manual Illinois: ABBOTTBASE, 1991; 6.7.1.
