The oxalate content of urine is believed to play an important role in the formation of calcium oxalate kidney stones1. As the concentration of calcium in urine is normally much higher than that of oxalate, small changes in oxalate concentration have much greater effects on calcium oxalate crystallization than changes in calcium concentration. Urinary oxalate is derived predominantly from two major sources: liver synthesis and the absorption of dietary oxalate. Ascorbate breakdown was once thought to make a significant contribution; however, this has been questioned, and the previous results are believed to reflect ascorbate breakdown during urine processing2,3. The contribution of the diet was thought to be low, in the order of 10 to 20%4, but the derivation of this contribution was based on studies with poor dietary control, older, less accurate oxalate assays, and an assumption that the absorption of intestinal oxalate was linearly related to the amount of oxalate ingested. Recent studies in our laboratory examining the ingestion of diets of known oxalate content suggest that the mean contribution of dietary oxalate to urinary oxalate is much higher (45%), that the relationship of dietary oxalate to the amount absorbed is not linear, and that it varies tremendously between individuals with a range of 10 to 72% (Holmes and Assimos, unpublished observations)5.
The main sources of dietary oxalate are plants and plant products, principally seeds and leafy plants related to spinach and rhubarb. In contrast, negligible amounts of oxalate occur in foods of animal origin. The physiological role of oxalate in plants is not precisely known. It has been suggested that it is involved in seed germination, calcium storage and regulation, ion balance, detoxification, structural strength, and insect repulsion6,7. At one extreme, the consumption of large amounts of plant oxalate and its absorption can be fatal to both humans and other animals because of oxalosis, the formation of calcium oxalate deposits in body tissues8,9. The ingestion of more moderate amounts of oxalate appears to play an important role in calcium oxalate kidney stone disease because of its absorption and excretion in urine3.
Estimates of the oxalate content of foods vary widely because of the inaccuracy of the analytical techniques that have been used. Previous methods have included colorimetric techniques, enzymatic procedures, and some types of chromatography3. Direct chromatographic techniques would appear to be ideally suited for the oxalate analyses of a wide variety of foods. Two such techniques, ion chromatography (IC) and capillary electrophoresis (CE), have been used in preliminary assays of foods1,3 and have been successfully used to measure urinary oxalate10,11,12. In this report, sample preparations for these two techniques have been developed, and the performance characteristics of the two assays have been compared in assaying the oxalate content of foods. These assays have been used to determine the mean oxalate intake of a small number of individuals.
METHODS
Materials
All chemicals were of analytical grade and were purchased from Sigma (St. Louis, MO, USA) or Aldrich (Milwaukee, WI, USA). Foods were obtained at a local supermarket.
Ion chromatography
The IC system consisted of a Waters 501 pump, a Rheodyne 7125 injector, a Waters 431 conductivity detector, an Alltech ERIS conductivity suppressor, and an Alltech All-Sep (10 cm) anion exchange column. The mobile phase used consisted of 0.85 mmol/L NaHCO3, 0.9 mmol/L Na2CO3 at a flow rate of 1.2 mL/min.
Capillary electrophoresis
A Waters Quanta 4000 was used with an electrolyte of 10 mmol/L Na2CrO4/0.5 mmol/L tetramethyltriammonium bromide and a negative power supply. Separations were obtained on a 60 cm (75 
m i.d.) fused silica capillary following a 100-second hydrostatic load and running at a constant current of 25 A. CE separates anions in the capillary based on their migration rate with the applied current and their interaction with the capillary wall. Indirect absorption at 254 nm is used to detect anions as they displace chromate in the electrolyte as it passes through the detection window.
Sample preparation
If required, food samples were chopped into small pieces and thoroughly mixed before 2 to 3 g was weighed into a 50 mL propylene screw cap centrifuge tube and mixed with nine volumes of 0.2 mol/L H3PO4 for CE assays or 0.2 mol/L HCl for IC assays. To ensure that samples had sufficient Cl- to stabilize migration times in CE analyses, the H3PO4 acid contained 800 mg NaCl/L. Foods containing very low oxalate were homogenized in 0.1 mol/L H3PO4 containing 400 mg NaCl/L to increase the sensitivity of the assay. Samples were homogenized for 30 seconds using a Polytron blade homogenizer. The homogenates were heated at 60°C for one hour to ensure complete solubilization of crystalline oxalate. Increasing the temperature or incubation time did not increase the oxalate extracted from several different food samples. Portions of the homogenate (1.5 mL) were removed for centrifugation in a microfuge (15,000 
g) for five minutes. Supernatants were further clarified by filtration through a 25 mm diameter 0.2 m pore size PTFE filter (Supelco, St. Louis, MO, USA). For both IC and CE, samples were diluted 50- to 500-fold in water for analysis with the extent of the dilution depending on the oxalate and anion contents of the sample.
Food records
Five normal individuals (3 females and 2 males, mean age 28.7 years) weighed and documented food consumption for three days. Representative samples of each food were returned to the laboratory for oxalate analysis. Twenty-four–hour urine collections were also obtained on each the day food consumption was recorded.
Statistical analysis
All data are presented as the mean 
SD. The relationship between variables was tested by linear regression. A P value of less than 0.05 was considered significant.
RESULTS
Oxalate extraction from foods
Oxalate exists in plants in a crystalline form either as calcium oxalate or as a soluble anion. Solubilization of the oxalate in foods was investigated using a hot acid extraction following a thorough blade homogenization. Initially, spinach was investigated since the water insoluble fraction was reported to comprise 30 to 70% of the total oxalate13. If all of the spinach calcium (100 mg/100 g) were present as calcium oxalate, it would complex only one third of the spinach oxalate (750 mg/100 g). We found that the amount of oxalate extracted from different batches of spinach without acid varied from 82 to 100%. Results with rhubarb stems were more consistent with 51 2% of oxalate extracted without acid in four different batches. The effect of acid concentration on oxalate extraction from rhubarb is shown in Figure 1, where CE was used to measure oxalate values. The extraction was essentially complete with 0.2 mol/L acid when the pH declined to less than 1.5. Similar results were obtained when HCl was used as the extracting acid and IC was used for measurements. Different extractant acids had to be used for the two techniques. High levels of Cl-, which migrates before oxalate on CE, interfere with CE analyses of oxalate, whereas phosphate migrates after oxalate and does not interfere. With IC, Cl- elutes well before oxalate and does not interfere, whereas phosphate elutes closer to oxalate and interferes.
Figure 1.
Acid extraction of rhubarb oxalate. Ten-percent homogenates were prepared in the concentration of acid indicated. After heating for one hour at 60°C, samples were centrifuged and supernatants assayed for pH (
) and oxalate by capillary electrophoresis (CE; 
). The results represent the mean ( SD) of three different batches of rhubarb.
Linearity of response
Both IC and CE gave a linear response over a 20-fold concentration range for standards prepared in acid at a concentration similar to that used in extractions. Calibration curves for both CE and IC are shown in Figure 2 (r2 = 0.99, P < 0.001 for both curves). The lower limit of quantitation for food analyses, defined as a signal to noise ratio of 10, was 0.2 mg/100 g for both methods. For IC, this value could potentially be lowered by using gradient elution and an ion exchange column producing sharper peaks and more theoretical plates than the column used in these studies. The upper limit for both methods is affected by the high concentration of the chloride and phosphate anions associated with the extractant acids, which affected oxalate peak areas.
Figure 2.
Calibration curves for ion chromatography (IC; ) and capillary electrophoresis (CE; 
) showing the relationship of peak area to the oxalate concentration in a standardized solution.
Resolution of oxalate in foods with ion chromatography and capillary electrophoresis
In both methods, oxalate was well resolved from other anions. Pherograms illustrating the analysis of oxalate in apple, potato, and spinach are shown in Figure 3, and chromatograms in Figure 4 show the analysis of the same foods by IC.
Figure 3.
Pherograms illustrating the measurement of the oxalate content of selected foods by capillary electrophoresis. The apple extract (A) was diluted 1/50, the potato (B) 1/100, and the spinach (C) 1/500. The migration time of the apple oxalate was 4.2 minutes, the potato oxalate 4.0 minutes, and spinach oxalate 3.9 minutes.
Full figure and legend (14K)Figure 4.
Chromatograms illustrating the measurement of the oxalate content of selected foods by ion chromatography. The apple extract (A) was diluted 1/50, the potato (B) 1/100, and the spinach (C) 1/500. The retention time of the apple oxalate was 15.2 minutes, potato oxalate 13.7 minutes, and spinach oxalate 14.1 minutes.
Full figure and legend (14K)Accuracy and reproducibility
The accuracy and reproducibility of both IC and CE were assessed, and the results are shown in Table 1. Both methods are satisfactory except for the reproducibility of the CE analysis of apple, a food with barely detectable oxalate content. The performance of IC was better than CE, as noted by the better reproducibility. The reproducibility of the CE method improved as the oxalate content of foods increased, and the CV was 6.3% with grape jelly (2.1 mg/100 g). A comparison of the assays of several different foods by both methods indicated that they are highly correlated (r2 = 0.99, P < 0.001), as shown in Figure 5. The complex nature of food matrices appeared to influence the reproducibility of both methods because of the binding of food components to columns. This eventually led to column poisoning, as evident with longer migration times with CE and shorter retention times with IC.
Figure 5.
Correlation of the measurements of selected foods by capillary electrophoresis and ion chromatography. Foods included spinach, wheat bran, white bread, potato, peas, and apple.
Full figure and legend (6K)
Oxalate content of various foods
The oxalate content of many foods was found to vary. Broccoli varied from 0.3 to 13 mg/100 g, potatoes from 5.5 to 30 mg/100 g, and wheat bran from 58 to 524 mg/100 g. The variability in other foods is depicted by the range of values noted in Table 2. Sweet potatoes, black olives, tomato, and vegetable soup were the most variable. The mean oxalate content of foods with a moderate to high oxalate content, which were assayed by CE, is shown in Table 2. As the oxalate content of foods becomes available in the future, values will be posted at the Website http://www.ixion-biotech.com, where food values we have previously published are displayed.
Oxalate intake
Five participants maintained food records on self-selected diets while obtaining 24-hour urine collections. These foods were assayed for oxalate, and the amount of oxalate ingested was calculated and reported in Table 3. Intakes ranged from 44 to 351 mg/day, with a mean daily intake for the whole group of 152 mg/day. There was no correlation between the amount of oxalate ingested each day and the amount of oxalate excreted in urine (r = 0.12).
DISCUSSION
The results of these studies show that two direct analytical techniques, CE and IC, are both accurate and reproducible in estimating the oxalate content of foods. Furthermore, the results obtained by these methods were comparable. IC was analytically superior to CE in estimating low values of oxalate, as is evident by the high coefficient of variation obtained for the CE analysis of apple. It can be expected that another direct technique for analyzing oxalate in foods, gas-liquid chromatography, would also be suitable for the analysis of oxalate in foods, but it will require a derivatization step to make oxalate volatile, thus increasing the assay time. It has been used previously for the analysis of a select number of foods, and results have agreed well with those obtained by CE3,14,15. For the routine analysis of foods, CE appears to be the method of choice over IC for two reasons. Running costs are much lower for CE primarily because of the 100-fold differential in the cost of columns, which have a limited lifetime with food analyzes using either method. With CE, a higher throughput than IC is possible because of the faster analysis time. Errors in estimating the oxalate content of low oxalate foods such as apple by CE can be tolerated given that this analytical error is likely to be less than the variations in oxalate content because of biological factors and because the contribution of such foods to daily oxalate intake is likely to be very low.
The two main procedures used previously to measure food oxalate have been a colorimetric procedure developed by Zarembski and Hodgkinson16 and an enzymatic method developed by Kasidas and Rose17, and they were recently used by the University of California San Diego Clinical Research Center18. The colorimetric procedure is not specific measuring the oxalate obtained by an ether extraction and calcium precipitation followed by the reduction of the oxalate to glycolate and a reaction with chromotropic acid19. This method could be subject to interference from food components in the extraction, precipitation, reduction, or colorimetric measurement steps. The oxalate oxidase-based procedure could be affected by inhibitors of the enzyme contained in foods, and there is likely to be significant interference with the colorimetric estimation of H2O2 in the final step. The limitations of these procedures are clearly evident in detecting only small amounts of oxalate in potato: 2.3 mg/100 g with the colorimetric procedure, and 1.0 mg/100 g with the enzymatic procedure. These values compare with the mean values of 22.7 to 24.7 mg/100 g that we report in Table 1 and to the range of 5.5 to 30 mg/100 g that was detected in a number of different batches of potatoes.
The variability in the oxalate content of foods will influence the accurate assessment of the amount of oxalate ingested in diets. To date, we have observed a large variability in many fresh foods and even in some processed foods. Factors known to influence the oxalate content of plants include the plant variety, the developmental stage of the plant, the season, and growth conditions13. In the face of this variation, the use of average values would appear prudent until a better understanding of factors influencing this variability is obtained and its impact is determined. Variability caused by food processing, food preparation, and factors that influence the bioavailability of the food oxalate after its ingestion is another consideration.
Development of this method for the estimation of the oxalate content of foods will permit a more accurate assessment of the oxalate intake of individuals and its impact on urinary oxalate excretion and calcium oxalate stone formation. The dietary oxalate intakes that we estimated for five individuals on three different days suggest that intake will be highly variable in individuals consuming typical North American diets, as their intakes ranged from 44 to 351 mg/day. In individuals consuming a normal portion of an oxalate-rich food, such as spinach, intake may exceed 1000 mg/day. The lack of a correlation between oxalate intake and urinary oxalate excretion is not surprising. The oxalate content of the foods ingested will be only one of several variables that influence the amount of oxalate absorbed. Other factors will include the bioavailability of the oxalate ingested, the amounts of oxalate-binding cations, such as calcium and magnesium that are ingested, the inherited capacity to absorb oxalate20, the transit time in the small and large intestines, and the activity of oxalate-degrading bacteria in the large intestine3. Crude calculations of the anticipated calcium and oxalate concentrations in intestinal fluids suggest that these solutions are supersaturated with calcium oxalate and that the bulk of the oxalate present is in the crystalline form. The concentration of ionized oxalate may be the principal determinant of the amount of oxalate absorbed. Whether factors influencing intestinal oxalate absorption or whether differences in endogenous oxalate synthesis account for the lack of correlation between oxalate intake and oxalate excretion could not be determined in these studies, but warrants further investigation. Our preliminary results with oxalate-controlled diets suggest that the absorption of dietary oxalate is highly variable between individuals and that it can contribute up to 72% of the urinary oxalate excreted5. If modifying the oxalate content of the diet decreases urinary oxalate excretion and stone formation, it could become an important tool in decreasing stone occurrence in stone formers.
References
- Robertson, WG & Hughes, H: Importance of mild hyperoxaluria in the pathogenesis of urolithiasis: New evidence from studies in the Arabian peninsula. Scanning Microsc 1993 7:391–402, | PubMed | ISI | ChemPort |
- Wandzilak, TR, D'andre, SD, Davis, PA, Williams, HE: Effect of high dose C on urinary oxalate levels. J Urol 1994 151:834–837, | PubMed | ISI | ChemPort |
- Holmes, RP, Goodman, HO, Assimos, DG: Dietary oxalate and its intestinal absorption. Scanning Microsc 1995 9:1109–1120, | PubMed | ISI | ChemPort |
- Williams, HE & Smith, LH: Primary hyperoxaluria, in The Metabolic Basis of Inherited Disease, 1983, edited by Stanbury JB, Wyngaarden JM, Fredrickson DS, Goldstein JL, Brown SMS, New York, McGraw-Hill, pp 204–208
- Holmes, RP, Goodman, HO, Assimos, DG: Metabolic effects of an oxalate-free diet, in Urolithiasis 1996, 1996, edited by Pak CYC, Resnick MI, Preminger GM, Dallas, Millet the Printer, pp 167–168
- Horner, HT & Wagner, BL: Calcium oxalate formation in higher plants, in Calcium Oxalate in Biological Systems, 1995, edited by Khan SR, Boca Raton, CRC Press, pp 53–72
- Lane, BG: Oxalate, germin, and the extracellular matrix of higher plants. FASEB J 1994 8:294–301, | PubMed | ISI | ChemPort |
- Sanz, P & Reig, R: Clinical and pathological findings in fatal plant oxalosis. Am J Forensic Med Pathol 1992 13:342–345, | PubMed | ISI | ChemPort |
- Allison, MJ, Daniel, SL, Cornick, NA: Oxalate degrading bacteria, in Calcium Oxalate in Biological Systems, 1995, edited by Khan SR, Boca Raton, CRC Press, pp 131–168
- Holmes, RP: Measurement of urinary oxalate and citrate by capillary electrophoresis and indirect ultraviolet absorbance. Clin Chem 1995 41:1297–1301, | PubMed | ISI | ChemPort |
- Menon, M & Mahle, CJ: Ion-chromatographic measurement of oxalate in unprocessed urine. Clin Chem 1983 29:369–371, | PubMed | ISI | ChemPort |
- Hagen, L, Walker, VR, Sutton, RA: Plasma and urinary oxalate and glycolate in healthy subjects. Clin Chem 1993 39:134–138, | PubMed | ISI | ChemPort |
- Libert, B & Franceschi, VR: Oxalate in crop plants. J Agric Food Chem 1987 35:926–938, | Article | ISI | ChemPort |
- Brinkley, L, McGuire, J, Gregory, J, Pak, CY: Bioavailability of oxalate in foods. Urology 1981 17:534–538, | Article | PubMed | ISI | ChemPort |
- Brinkley, LJ, Gregory, J, Pak, CYC: A further study of oxalate bioavailability in foods. J Urol 1990 144:94–96, | PubMed | ISI | ChemPort |
- Zarembski, PM & Hodgkinson, A: The oxalate content of English diets. Br J Nutr 1962 16:672–681,
- Kasidas, GP & Rose, GA: Oxalate content of some common foods: Determination by an enzymatic method. J Hum Nutr 1980 34:255–266, | PubMed | ISI | ChemPort |
- Brezinski, E, Durning, AM, Grasse, B, Fusselman, E, Ciaraldi, T: Oxalate Content of Selected Foods. 1996 San Diego, University of California
- Hodgkinson, A & Zarembski, PM: The determination of oxalic acid in urine. Analyst 1961 86:16–21, | Article | ISI | ChemPort |
- Holmes, RP, Assimos, DG, Goodman, HO: Genetic and dietary influences on urinary oxalate excretion. Urol Res 1998 26:195–200, | Article | PubMed | ISI | ChemPort |
Acknowledgments
This research was supported by NASA grant NAG 5-3968. The technical assistance of Ms. Felicia Russell was greatly appreciated.
