CLIN.CHEM. 28/7,1457-1460(1982) CLINICAL CHEMISTRY, Vol. 28, No. 7, 1982 1457 Simple, ReliableChromatographicMeasurementof OxalateinUrine Antonio Di Corcia,’ Roberto Samperi,1 Giuliana Vinci,1 and Giuseppe D’Ascenzo2 In this assay foroxalateinurine,oxalateisadsorbedfrom the urineonto graphitizedcarbon black(Carbopack B). Afterdesorptionand removal ofthe solvent,oxalicacid is gas-chromatographically measured after being deriva- tizedwithBF3/methanol.Chromatography is on Carbopack B/polyethylene glycol (M. 20 000),93.7/6.3byweight.The lower limit of detection of urinary oxalate is about 3 mg/L (CV 3.6%). A seriesofoxalatedeterminationsin24-h urinesamples of 15 subjectsgave a mean of 47.1 (SD 15.4) mg/24 h, with an analytical recoveryof97.4% (SD 3.0%, range 93.6-102.4%). Total analysis time for one sample is about 2 h. AddItIonal Keyphrases: adsorption desorption den- vatization #{149} gas chromatography Determination of urinary oxalate is of much interest for the diagnosis of various types of primary hyperoxaluria. Knowl- edge of the oxalate content of urine from patients with calcium oxalate nephrolithiasis, correlated with other data, can also aid urologists in clarifying which biochemical and (or) physio- logical mechanisms are operative in stone formation. Although there has been considerable work on methods for urinary oxalate measurement, almost none of these combine simplicity, rapidity, and reliability. Enzymic methods (1,2) are highly sensitive and require no sample pretreatment but are time consuming and expensive (3). Colorimetric determinations (4) are simple to perform but are affected by several positively interfering substances that can cause overestimation of urinary oxalate (5). Gas chromatography (GC), being both specific and sensi- tive, has been extensively exploited for determination of ox- alate in both urine and serum (5-7). Nevertheless, many of the GC methods reported involve time-consuming sample- purification procedures, such as precipitation. Adsorbing materials for cleanup of biological samples are increasingly attracting attention for determination of com- pounds of biomedical interest (8-10). Column techniques for this are rapid, reliable, and simple, and offer additional se- lectivity when combined with GC analysis. Here we describe an accurate, rapid, and simple method for determining oxalate in urine. Samples are passed through a colunm filled with graphitized carbon black as adsorbent. The retained oxalate is eluted with acidified methanol. This sol- vent is then evaporated and the residue, derivatized with BF3 in methanol, is measured by GC. Materials and Methods Materials. We obtained 24-h urine specimens from 15 subjects, with 30 mL of toluene in the container as preserva- Istituto di Chimica Analitica, Universit#{224} Degli Studi di Roma, 00185 Roma, Italy. 2 Istituto Chimico, Universit#{224} di Camerino, Italy. Received Feb. 8, 1982; accepted April 6, 1982. tive. Oxalic acid (99.5%), 2-heptanone (the internal standard), and boron trifluoride (100 mL/L of methanol) derivatizing agent were all obtained from Fluka AG, Switzerland. Gra- phitized carbon black (Carbopack B, 80-120 mesh) was from Supelco Inc., Bellefonte, PA 16823. Polyethylene glycol (Mr 20 000) was from Carlo Erba, Milan, Italy. We used the Carlo Erba Model GI gas chromatograph equipped with a flame ionization detector, and we quantified peaks with a Shimadzu Model Chromatopac C-E 1 B inte- grator. The gas-chromatographic column adopted for quan- titation was glass, 2m X 2 mm (i.d.), packed with the same material used for the cleanup procedure-that is, Carbopack B-suitably modified with added polyethylene glycol (PEG). Preparation of the packing material and the column packing procedure were as previously described in detail (11, 12); final proportions of materials are Carbopack/PEG 93.7/6.3 by weight. The GC column was conditioned overnight at 250 #{176}C with nitrogen. For the quantitation procedure, the column was operated at 145 #{176}C with nitrogen as carrier gas and with a dead time of 25 s. The injection-port temperature was 180 #{176}C. A 5-sL Hamilton syringe, Model 801, was used to inject the final sample. Procedure. Prepare the purification column by suspending 0.6 g of Carbopack B particles in 0.1 mol/L aqueous HC1 and introducing the suspension into a 14 X 0.6 cm glass column fitted with a Teflon stopcock. Pack the adsorbent by tapping the column while passing distilled water through it, until the column of carbon is 6.5cm high and the effluent water is at pH 6 (pH paper). Pass through the column, at a flow-rate of about 0.6 mL/min, 2 mL of the 24-h urine, previously diluted with an equal volume of water and then acidified with concentrated HC1 to pH 3 (pH meter). When the meniscus of the sample reaches the top of the carbon, wash the column by passing through 1.5 mL of water acidified with HC1 to pH 3. Use methanol acidified with HC1 (0.1 mol/L) to desorb oxalic acid from Carbopack B. Discard the first 4.6 mL of the effluent from the column and collect the following 1.3 mL, which contains oxalic acid, in a 10-mL air-tight screw-capped glass vial fitted with a silicone-rubber septum. The discarded fraction consists of 1.2 mL of dead volume, 2 mL of diluted urine, and 1.4 mL of the acidified water just following the urine volume. Evaporate the collected fraction-mainly methanol, with some water-under a stream of nitrogen while the tube is in a water bath maintained at 50 ± 1 #{176}C. Derivatize the residue by adding 0.1 mL of the methanolic BF3, capping the vial, and incubating for 5 mm in a water bath at about 70 #{176}C. After cooling the tube, add 10 L of 2-heptanone solution (1.11 gIL), then inject 0.6 tL onto the GC column. Figure 1A shows a typical GC profile for an unsupple- mented urine. Calculations. To calculate the amount of oxalate in urine from the GC peak, prepare an oxalic acid/water/methanolic HC1 standard solution and evaporate a known volume of it. After derivatization, add 10 sL of the 2-heptanone standard solution, make replicate injections, and measure the peak area ratios of the dimethyl oxalate and the internal standard. We