B-203 Development and Validation of an LC/MS/MS Method to Quantify Creatinine in Dried Blood Spots

Abstract

Abstract Background Health disparities in diagnosis and treatment of chronic kidney disease (CKD) are apparent with Black, Brown, and socially disadvantaged people experiencing worse outcomes. Improving access to CKD screening may decrease disparities. Blood creatinine concentrations are a mainstay in CKD screening, as it is commonly used to estimate glomerular filtration rate. Creatinine is classically measured in serum/plasma using high-throughput chemistry analyzers. Dried blood spots (DBS) could offer a solution to under-screening, but require an alternative method since extraction from DBS retrieves minimal sample volume and is difficult to scale using conventional automation. The purpose of this study was to develop an LC/MS/MS DBS method to quantify creatinine and ultimately to increase screening access for underprivileged communities. Methods Creatinine was measured using a 100 × 2.1 mm Hypersil Silica column with an ABSciex API4000 MS/MS detector. A 5-point calibration curve was made using creatinine certified reference material (Cerilliant) and blank whole blood (UTAK) (target concentrations: 0.575, 0.975, 4.575, 10.575, 20.575 mg/dL). Calibrators were spotted onto Whatman903 cards and allowed to dry for at least 24-hours. A 6 mm punch was used for analysis of calibration points and patient samples. Once punched, water (50 ul) was added and the samples were incubated at 60°C for 15 min before shaking at 800 RPM. Acetonitrile with 2H-creatinine internal standard (IS; .06 mg/dL) was added and the mixture was shaken and centrifuged. The resulting supernatant was diluted to contain 20% water before injecting 10 uL. Creatinine was separated using an isocratic mobile phase (80% Acetonitrile, 20% DI water, 10 mM Ammonium Formate) and detected using MRM mode (Q1 114.0 m/z; Q3 86.8 m/z (quantifier) and 44.5 m/z (qualifier)). For the IS a Q1 of 117.0 m/z and Q3 of 89.9 m/z was applied and the extraction ratio of endogenous creatinine to IS was calculated to determine the concentration. Linearity was confirmed by running five levels of DBS calibration standards for five days. Imprecision was established by running 3 control levels five times for five consecutive days. Control material was purchased from BioRad and run as liquid. Accuracy was assessed using residual liquid CAP serum creatinine proficiency testing (PT) material in triplicate. DBS creatinine concentration determined by LC/MS/MS was compared to paired venous serum creatinine concentration determined by the Roche Cobas enzymatic creatinine assay (n = 63). Results Creatinine concentrations from 0.2–20 mg/dL were within the allowable nonlinearity of 0.075 mg/dL or 5.0% (y = 1.00x−0.010;R = 0.999). Intraday and interday precision was <2%and <3%, respectively for all levels (mean concentrations 1.00, 2.02, and 7.05 mg/dL). Residual CAP PT samples (n = 4) were within 2.5% of the peer group mean. Compared to venous results, DBS had a bias of −0.03 mg/dL (−3.573%) and a linear equation of (y = .904x +0.049; R = 0.83). Conclusion Preliminary results indicate that DBS are a viable sample source for measuring creatinine. Creatinine concentrations measured by DBS compared favorably with CAP PT. A clinically insignificant negative bias was seen when DBS creatinine concentration was compared to venous blood.