Original Communication | Published:

Bioavailability of soluble oxalate from tea and the effect of consuming milk with the tea

European Journal of Clinical Nutrition volume 57, pages 415419 (2003) | Download Citation



Objectives: To measure the availability of oxalate normally extracted when making tea from two commercially available black teas bought from a supermarket in Christchurch, New Zealand in July 2001.

Design, subjects and intervention: A randomized double crossover study. Six students and four staff consumed six cups of each brand of tea both with and without added milk over a 24 h period. A total urine collection was taken for the initial 6 h followed by a further 18 h. The oxalate content of the urine voided was measured using an enzyme kit method and the availability of the soluble oxalate consumed was measured for the 6 h and the total 24 h sample.

Setting: University campus.

Results: The mean soluble oxalate content of black tea in the two different commercial tea bags was respectively 6.1 and 6.3 mg soluble oxalate/g tea. The mean availability of the oxalate extracted from tea measured over a 6 h period ranged from 1.9 to 4.7% when tea was consumed without milk. The availability of the soluble oxalate from tea ranged from −3.0 to 2.3% for each of the two brands of tea investigated over a 24 h period.

Conclusion: These studies show that consuming black tea on a daily basis will lead to a moderate intake of soluble oxalate each day, however the consumption of tea with milk on a regular basis will result in the absorption of very little oxalate from tea.

Sponsorship: Lincoln University.


A high oxalate uptake from the diet is thought to play a role in hyperoxaluria, a documented risk factor in the formation of calcium oxalate kidney stones (Noonan & Savage, 1999). Absorptive or ‘dietary’ hyperoxaluria is generally thought to be a direct consequence of oxalate bioavailability. Therefore, people with an increased risk for calcium oxalate stone formation are commonly advised to avoid consuming oxalate-rich foods. A number of foods such as spinach, rhubarb, beets, nuts, chocolate, wheat bran and strawberries are known to contain high oxalate levels (Noonan & Savage, 1999). These are foodstuffs that have a low ratio of oxalate to calcium and are not thought to have a major effect on calcium availability. Although tea is reported to contain between 300 and 2000 mg oxalate/100 g fresh weight, it is not consumed in the same quantities or in the same way as other foods. In many cases it is taken with milk, which reduces its oxalate/calcium ratio. In some publications it is not clear whether the oxalate content has been measured in tealeaves or in brewed tea. More recently Charrier et al (2002) have shown that tea sold loose or in tea bags contains between 1.5 and 6.8 mg soluble oxalate/g tea. They went on to show that the consumption of six cups of tea made from either loose black tea or black tea in tea bags would result in an intake of between 16 and 102 mg soluble oxalate/day. These levels are modest when compared with the amounts of soluble oxalate found in common foods (Noonan & Savage, 1999; Savage et al, 2000; Albihn & Savage, 2001a). Tea drinking, however, tends to occur on a daily basis, confirming the observation made by Zarembski and Hodgkinson (1962) that ‘tea is a significant source of oxalate intake in English diets’.

Consumption of tea without milk has been shown to increase urinary oxalate concentrations (Brinkley et al, 1990; Finch et al, 1981). These findings have prompted a recommendation to eliminate black tea from the diets of those people who form oxalate stones (Massey et al, 1993). McKay et al (1995) showed that steeping tea for different times and stirring the tea had a significant effect on the oxalate content of the final extraction. From their data a regular tea drinker consuming six cups of tea per day would consume between 65 and 124 mg soluble oxalate per day, which is comparable to the results obtained in the study carried out by Charrier et al (2002). If black tea is taken with added milk or is consumed with calcium-containing meals then less oxalate will be absorbed as some soluble oxalate extracted from the tea leaves will bind to calcium and continue down the digestive tract without being absorbed (Massey, 2000).

Oxalate absorption from tea can elevate urinary oxalate levels, possibly leading to an increased risk of kidney oxalate stone formation. However two prospective studies show that the consumption of tea actually leads to a reduced risk of kidney stone formation by 8% in females (Curhan et al, 1998) and 14% in males (Curhan et al, 1996). Two possible positive benefits of tea consumption are its antioxidant content and the increase in urinary volume tea drinking produces (Massey, 2000), thus reducing the risk of stone formation even when oxalate excretion is high.

Most studies show a peak in oxalate absorption after 1–6 h in normal subjects following ingestion of oxalate containing food (Barilla et al, 1978; Brinkley et al, 1981, 1990; Marangella et al, 1982; Liebman & Chai, 1997). This indicates that there is a major uptake of oxalate from the small intestine in healthy humans. The relationship between soluble and insoluble oxalate in the small intestine seems to have a major effect on oxalate bioavailability, since ingestion of calcium together with oxalate-rich foods has been shown to lower the uptake of both calcium and oxalates (Marshall et al, 1972; Barilla et al, 1978; Heaney et al, 1988; Masai et al, 1995; Holmes et al, 1996; Weaver et al, 1997; Savage et al, 2000). This indicates that insoluble calcium oxalate has a much lower bioavailability than the soluble from of oxalate, and that an oxalate rich/low calcium diet leads to a greater uptake of oxalate. The fact that brewed tea only contains soluble oxalates might place it in the group of high-risk foods for people predisposed to renal oxalate calculi formation, since soluble oxalate appears to have a higher bioavailability.

This study was carried out to measure the soluble oxalate content of two popular teas currently available in New Zealand and to determine the bioavailability of oxalate when consumed both with and without milk.


Sampling and extraction

Two samples of black tea were bought from supermarkets in Christchurch in June 2001. Each sample of tea in a teabag was infused in 245 ml of hot (90°C) nanopure water for 5 min. Each tea was stirred twice and then the tea bag was removed and squeezed twice. The tea infusions were cooled to room temperature and 20 ml of the tea solution was transferred to 100 ml volumetric flask and made up to the volume with nanopure water. This procedure extracts the soluble oxalates from the tealeaves; the insoluble oxalates remain in the discarded tealeaves.

Soluble oxalic acid determination

Each sample of tea was filtered through a 0.45 mm cellulose acetate membrane syringe filter (Sartorius, Germany) and analysed in triplicate using the method outlined in detail by Savage et al (2000) based on an earlier method (Holloway et al, 1989). Five microlitres of filtered sample was injected onto a Waters high pressure liquid chromatograph (HPLC) with a UV detector set at 210 nm. The chromatographic separation was carried out on a 300 × 7.8 mm Biorad Aminex Ion exclusion column (HPX-87H) attached to an Aminex cation guard column using an isocratic elution at 0.5 ml/min with 0.0125 M sulphuric acid (Analar, BDH, UK) as the mobile phase. The analytical column was held at 50°C. The oxalic acid peak was identified by comparison of the retention time with a range of common plant organic acid standards. The amount of oxalic acid in each sample was determined by using external standards. The data are presented as the mean of three determinations.

Experimental design

Ten healthy non-stone-forming subjects were recruited from staff and students at Lincoln University, Canterbury, NZ to a randomized double-crossover study. After informed consent was obtained, the volunteers, five women and five men (21–54 y old) participated in the study. Each volunteer consumed six cups of tea per day at 9.00 a.m., 10.30 a.m., 12.00 noon, 2.30 p.m., 4.00 p.m. and 6.00 p.m. The tea was prepared in the same way as the earlier extraction method. After consuming the first cup of tea, a total 6 h urine sample was commenced on each volunteer. A second container was given to each volunteer at 3 p.m. to collect the urine for the following 18 h. In the first experiment each volunteer drank six cups of tea without milk, in the second experiment the same tea was consumed with 25 ml low-fat milk (Meadow Fresh, green top, Mainland Products Ltd, Christchurch, New Zealand), which contained 40 mg calcium. These two experiments were repeated for the second brand of tea. Volunteers were allowed to add some sugar to sweeten their tea as required. In a final experiment the volunteers were required to drink water in place of tea; urine was collected in the same way as for the tea drinking experiment.

During the experiment the volunteers were asked to keep their normal diet, and not to increase or decrease their calcium intake and to avoid oxalate-rich foods like spinach, rhubarb, tea and silvo beet, the day before each experiment and the day of the experiment and during the urine collection. They were also asked to avoid consuming extensive amounts of medium oxalate-containing food such as peanuts, chocolate and strawberries, but no control of the oxalate intake was undertaken in this study. The volunteers consumed a continental breakfast at 8 a.m., a light lunch at 12.00 noon and an evening meal at 7 p.m.

The bioavailability of oxalate was calculated by expressing the output of oxalate in the urine as a percentage of the oxalate consumed in tea over the same time period.

Urine collection and sampling

Urine was collected into plastic containers and 10 ml concentrated HCL was added to each sample immediately after the urine was collected. The total volume of each container was measured and 100 ml samples were taken and analysed by Canterbury Health Laboratories, Christchurch, NZ, using a Sigma oxalate kit (no. 591-D).

Statistical analysis

Significant differences in the mean output of oxalate in the urine were determined using a Generalized Linear Model Procedure (Minitab 11).


The total soluble oxalate extracted from black tea in teabags was 6.07 and 6.22 mg/g tea, respectively. As the teabags contained a mean of 2.5 g of tea, a cup of tea would contain 15.1 and 15.6 mg of oxalate and therefore the consumption of four cups of each tea would, respectively, lead to an intake of 60.4 and 62.4 mg oxalate during the 6 h test period. The output of oxalate in the 6 h period when the four cups of tea were drunk without milk led to an output of between 6.4 and 8.0 mg oxalate (Tables 1 and 2). The output of oxalate in the urine in the initial 6 h period fell to 3.7 and 5.5 mg oxalate when the same two teas were consumed with 25 ml low-fat milk.

Table 1: Mean intake of oxalate from tea and urinary oxalate excretion (mg; ± s.e.) for 0–6, 6–24 and 0–24 h following consumption of tea with and without added milk
Table 2: Main effect and interaction significant levels (P-values) for the consumption of tea and tea with milk analysis

The apparent recovery of oxalate in the urine following consumption of four cups of tea during the initial 6 h period using the method of Albihn and Savage (2001a) ranged from a mean of 11.7% for both of the black teas to a mean of 7.5% when the same teas were consumed with milk. Output of oxalate in the urine over 6 h was significantly reduced (P<0.001) when milk was consumed with the tea. The mean output of oxalate in the urine over 6 h when no tea was consumed was 5.2±0.5 mg oxalate.

The availability of oxalate consumed from the black tea over 6 h ranged from 1.9 to 4.7% for the two tea brands, when corrected for the output of oxalate when no tea was consumed. When allowance for the background flow of oxalate is taken from the mean output of oxalate following the consumption of tea with milk during the 6 h period, effectively no oxalate in the urine appears to be derived from the tea although there are differences between the brands of tea.

Over the 24 h period between 28.9 and 24.0 mg oxalate was voided by the volunteers consuming the black tea following a total intake of between 90.7 and 93.7 mg oxalate from the tea. In all cases the output of oxalate in the urine over 24 h was significantly reduced (P<0.001) when milk was consumed with the tea.

The availability of oxalate from the black tea over 24 h ranged from−3.0 to 2.3% for the two tea brands, when the output of oxalate was corrected for the output of oxalate when no tea was consumed. Over 24 h there was a highly significant (P<0.01) reduction in oxalate output when milk was consumed with the two teas.


Charrier et al (2002) showed that a regular tea drinker consuming six cups of tea/day would have an intake of between 17.9 and 93.7 mg soluble oxalate/day from black tea in tea bags. While this intake of oxalate would occur on a daily basis these amounts are modest when compared to the soluble oxalate content of some common foods (Noonan & Savage, 1999; Albihn & Savage, 2001b). In the present experiment a mean of 92.2 mg oxalate was consumed per day and these results confirm the observation made by Zarembski and Hodgkinson (1962) that ‘tea is a significant source of oxalate intake in English diets’.

In this experiment no rigorous control of the diet was maintained, the volunteers were simply asked to keep to their normal diets but refrain from consuming oxalate rich foods for the day before and during the experiment. The low variability of the mean reference oxalate output (Table 1) confirms that none of the volunteers were stone formers as these people would excrete significantly higher levels of oxalate in their urine (Marangella et al, 1982). The low variability of the mean oxalate output suggests that the volunteers avoided other oxalate containing foods as requested.

The availability of oxalate consumed from the two tea brands calculated over the 6 h collection period ranged from 1.9 to 4.7% when the output of oxalate in urine was corrected for the output of oxalate when no tea was consumed. These values are comparable to previously determined availability of soluble oxalates from other foods for instance 2.4% (Brinkley et al, 1981) and oca 2.4% (Albihn & Savage, 2001b). In contrast, earlier reports on the availability of oxalate from lower intakes of oxalate from tea report an availability of 22.2% calculated from a 24 h collection on three subjects (Finch et al, 1981) while Brinkley et al (1990) determined the availability calculated over an 8 h period from a single ingestion of 1.4 g oxalate from tea to be 0.08%.

In this experiment it is clear that the consumption of milk significantly reduces (P<0.001) the subsequent output of oxalate in the urine in the 0–6 h and the 0–24 h periods, presumably by calcium in the milk binding to the soluble oxalate extracted from the tea. It is interesting to note that, when the mean difference in the oxalate output following consumption of tea with milk is compared to the oxalate output when no tea was consumed (the reference value), a negative value was observed for all three time periods considered.

It is reasonable to suggest that the soluble oxalate in the tea is binding with calcium in the milk and not being absorbed in the intestinal tract. Charrier et al (2002) showed that, if 25 ml of non-fat milk (containing 160 mg Ca/100 ml) was consumed with the same two brands of tea, then the calcium in the milk would have the capacity to bind to between 168 and 177% of the oxalate in the two teas.


This study shows that regularly consuming tea will lead to a moderate intake of soluble oxalate each day. It would, however, be very difficult indeed to consume an excessive amount of soluble oxalate from drinking tea. The consumption of tea with milk would mean that very little oxalate would be absorbed from tea on a daily basis and would place tea in the low-risk group of foods as defined by Noonan and Savage (1999).


  1. & (2001a). The effect of cooking on the location and concentration of oxalate in three cultivars of New Zealand-grown oca (Oxalis tuberosa Mol). J. Sci. Food Agric., 81, 1–7.

  2. & (2001b). The bioavailability of oxalate from oca (Oxalis tuberosa). J. Urol., 166, 420–422.

  3. , , & (1978). Renal oxalate excretion following oral loads in patients with ileal disease and with renal and absorptive hypercalciurias, effect of calcium and magnesium. Am. J. Med., 64, 579–585.

  4. , , & (1981). Bioavailability of oxalate in foods. Urology, 17, 534–553.

  5. , & (1990). A further study of oxalate bioavailability. J. Urol., 144, 94–99.

  6. , & (2002). Oxalate content and calcium binding capacity of tea and herbal teas. Asia Pacific J. Clin. Nutr., (in press)

  7. , & et al (1996). Prospective study of beverage use and the risk of kidney stones. Am. J. Epidemiol., 143, 240–247.

  8. , , & (1998). Beverage use and risk for kidney stones in women. Ann. Intern. Med., 128, 534–540.

  9. , & (1981). Urine composition in normal subjects after oral ingestion of oxalate-rich foods. Clin. Sci., 60, 411–418.

  10. , & (1988). Calcium absorbability from spinach. Am. J. Clin. Nutr., 47, 707–709.

  11. , & (1996). Metabolic effects of an oxalate-free-diet. In:Urolithias, ed. CYC Pak, M I Resnick & GM Preminger, pp.167–168, Dallas, TX: Millet

  12. & (1997). Effect of dietary calcium on urinary oxalate excretion after oxalate loads. Am. J. Clin. Nutr., 65, 1453–1459.

  13. , , & (1982). Hyperoxaluria in ideopathic calcium stone disease: further evidence of intestinal hyperabsorbtion of oxalate. Clin. Sci., 63, 381–385.

  14. , & (1972). Relationships between calcium and oxalic acid intake in the diet and their excretion in the urine of normal and renal-stone-forming subjects. Clin. Sci., 43, 91–99.

  15. , & (1995). Effect of dietary intake on urinary oxalate excretion in calcium renal stone disease. Br. J. Urol., 76, 692–699.

  16. (2000). Tea oxalate. Nutr. Rev., 58, 88–89.

  17. , & (1993). Effect of dietary oxalate and calcium on urinary oxalate and risk of formation of calcium oxalate kidney stones. J. Am. Diet. Assoc., 93, 901–906.

  18. , , , & (1995). Herbal tea: An alternative to regular tea for those who form calcium oxalate stones. J. Am. Diet. Assoc., 95, 360–361.

  19. & (1999). Oxalate content of foods and its effect on humans. Asia Pacific J. Clin. Nutr., 8, 64–67.

  20. , , & (2000). Effect of cooking on the soluble and insoluble oxalate content of some New Zealand foods. J. Food Comp. Anal., 13, 201–206.

  21. , , & (1997). Calcium bioavailability from high oxalate vegetables: Chinese vegetables, sweet potatoes, potatoes and rhubarb. J. Food. Sci., 62, 524–552.

  22. & (1962). The oxalic content of English diets. Br. J. Nutr., 16, 627–663.

Download references


The authors wish to acknowledge Trevor Walmsley, Canterbury Health Laboratories, Christchurch, NZ, for his assistance with the analysis of urinary oxalate and Professor D McNeil, Plant Sciences, Lincoln University for his assistance with the statistical analysis of the data.

Author information


  1. Food Group, AFSD, Lincoln University, Canterbury, New Zealand

    • G P Savage
    •  & L Vanhanen
  2. Institut des Sciences et Technics, Engenierieur d'Agers, Universite d'Angers, Angers, France

    • M J S Charrier


  1. Search for G P Savage in:

  2. Search for M J S Charrier in:

  3. Search for L Vanhanen in:


Guarantor: GP Savage.

Contributors: MJSC carried out the study, LV provided technical support, GPS supervised the study and wrote the manuscript.

Corresponding author

Correspondence to G P Savage.

About this article

Publication history







Further reading