Tuesday, July 4, 2017

Fluoride in tea: good or bad; how much is too much?


Someone asked this question on Steepster, if fluoride in tea is a good thing or a concern.  I took a look around about that, and I'll use a blog post to share what turned up.

It comes up that taking in some amounts of fluoride could be too much, and as with everything else discussion of what level is a health risk varies.  So it seems that the fluoride in tea would either be a very good thing or a bad thing depending on how much fluoride someone is already taking in, perhaps mostly through treated water.  Lets get to reviewing that.


The approach here is to review the following:


-what is the recommended intake range of fluoride, and how much would be too much?

-how much fluoride would people ingest through that added to municipal water?  I don't think it's added here in Bangkok, where I live, but I'll still look into that.

-how much is in different types of tea?

-where does that all leave us, is it probably good or not, or is there potential risk in some cases?


the happy side of fluoride (image credit)


Someone interested in the short version could skip ahead to the conclusion section, since all of this gets a bit complicated, and the conclusions summarize all the rest.



This post doesn't really focus on the effects of taking in too much fluoride, which I completely left out of a first version, but here is a relatively short description of that from an EPA reference document:


Adults exposed to excessive consumption of fluoride over a lifetime may have increased likelihood of bone fractures, and may result in effects on bone leading to pain and tenderness. For effects to teeth, children are most likely to be affected by excessive exposure to fluoride because it impacts teeth while they are still in formative phases. Children aged 8 years and younger exposed to excessive amounts of fluoride have an increased chance of developing pits in the tooth enamel, along with a range of cosmetic effects to teeth. For prevention of tooth decay, the beneficial effects of fluoride extend throughout the life span. 


So fluoride is a good thing, in the right amount.  In the first version of this post I didn't include any reference to how rare this condition is, but I'll add that here, cited from this research paper source:


In the USA, the prevalence of dental fluorosis appears to be increasing. In children aged 15–17 years, the 1999–2004 National Health and Nutrition Examination Survey (NHANES) found 40.6% had very mild or greater enamel fluorosis, up from 22.6% in the 1986–1987 study (fig. 3)...

The incidence of very mild and greater fluorosis in persons aged 6–39 years was 19.79% in white non-Hispanics, 32.88% in black non-Hispanics, and 25.8% in Hispanics (table 3). The increased prevalence of fluorosis in black non-Hispanics may suggest a genetic influence on fluorosis susceptibility.


In that paper cases of severe fluorosis are cited as the following percentages:




There seems a good chance that only "moderate / severe" cases might be recognizable related to the most negative effects, which on that table only relates to 1.81 to 4.03% of the people studied (which is still higher than I would have expected).


Research on how much fluoride is good or bad for you


A general medical site like WebMD is good for listing some basic starting point information, perhaps if not for serving as a final word on a subject, so let's check an excerpt there:




Clear enough?  I'll summarize:  they recommend you take in 3 mg. / day for women and 4 mg. / day for men, but not more than 10 mg. per day for either.  Those levels are much lower for children.

It's clear in that post that the risk relates to long term consumption levels, since it says that fluoride is used to treat osteoporosis at a dosage of 15 to 20 mg per day, up to double the limit for long term exposure.  It makes you wonder how that's going to work out related to other findings, the amounts in water, which should be easy to figure out given that typical range of .7 to 1.2 ppm.

I never do get around to reviewing the input of toothpaste here; that might be limited factor, even if you do spit it out.  It's interesting that fluoride is part of more than one compound, in that last citation sentence; that's something else I won't review.


A nice summary reference by the EPA covers similar range.  I'll check that against another better source of input (or one that might be better), a research paper entitled "Chronic Fluoride Toxicity: Dental Fluorosis," which lists how much fluoride is safe, it just takes some looking around to find it:


Prevention of Dental Fluorosis:  Dental fluorosis can be limited or prevented by following the ‘recommended limits for fluoride exposure’, suggested by US Environmental Protection Agency (USEPA) [20]. The reference dose suggested by USEPA is 0.06 mg fluoride/kg/ day, which is the estimate of daily exposure that is likely to be without any appreciable risk of deleterious effects (any degrees of dental fluorosis) during a lifetime...


I weigh around 70 kilograms (just under 160 pounds) and that comes out to a limit of 4.2 mg. per day.  That's quite a bit less than the 10 gm/day limit WebMD listed.


Let's check one of those health-risk-warning type sites (Fluoride Alert), and see if they're saying the same thing:


The current “safe” daily dose for fluoride fails to withstand scrutiny: The Institute of Medicine (IOM) states that anyone over 8 years of age — irrespective of their health condition — can safely ingest 10 milligrams of fluoride each day for their entire life without developing symptomatic bone damage. Ten milligrams, however, is the same dose that the IOM concedes can cause clinical signs of skeletal fluorosis within just 10 to 20 years of exposure...   

Some people are particularly susceptible to fluoride toxicity: It is well known that individual susceptibility to fluoride varies greatly across the population...

...The margin between the toxic and therapeutic dose is very narrow: The NRC concluded that the allegedly “safe” upper limit of fluoride in water (4 mg/l) is toxic to human health. 


If someone drank two liters of water per day that 4 mg/l would put them at 8 mg/day, which this is saying is toxic.  The main reason here seems to tie back to people's tolerance varying, more than that some people might manage to drink over 2 1/2 liters of treated water a day.

It's possible that both of these sources are interpreting risks on the higher side as assessed by conventional wisdom, but worth keeping in mind that that 4 mg/l level is still quite a bit over the .7-1.2 mg/l cited as a conventional range by WebMD (over three times that higher limit).  It had been cited as ppm there (parts per million) but they work out to be the same thing, based on the miracle of the metric system, and related to a milliliter of water weighing a gram.


Checking back in with that "Chronic Fluoride Toxicity" paper, lets see what levels they claim is in municipal water:


In the USA, approximately 10 million people are exposed to naturally fluoridated public water. In 1993, it was reported that 6.7 million people drank water with fluoride concentrations ≤1.2 mg/l, 1.4 million drank water with 1.3–1.9 mg/l fluoride, 1.4 million drank water with fluoride between 2.0 and 3.9 mg/l and 200,000 people ingested water with fluoride concentrations ≥4.0 mg/l [16]. Some areas have extremely high concentrations of fluoride in drinking water – such as in Colorado (11.2 mg/l), Oklahoma (12.0 mg/l), New Mexico (13.0 mg/l) and Idaho (15.9 mg/l) [9]... 


If that really is right drinking one liter of water in those last three states (in the right locations) would be too much fluoride.  It makes you wonder why that data is from 24 years ago; this paper was published in 2011.  One reason could be a comprehensive study is available from then, or they could be picking data that matches the highest risk, trying to discuss a worst case.  It's seems quite possible those samples relate to untreated sampled water, to well water and spring water sources, and if so in most cases people may still be drinking that same water that includes those same levels of fluoride.


WHO risk map of local water source fluoride levels (image source)



The research says 2/3rds of all water is treated to within that typical range (under 1.2), with three levels above that, although only 200,000 people are exposed to concentrations over 4 mg / liter.


This is going to be tricky to summarize related to tea input as a good or bad thing, since some people could already be taking in too much fluoride from a water source.  It will work better to say how much is fine for people not taking in fluoride from treated water, or how it works for people within that standard range.


If someone drank two liters of water per day (a bit over 8 cups, a standard amount for a total) treated to 1.2 mg per liter (the standard level) then they'd be at 2.4 milligrams intake per day.

That's perhaps not at risk yet, but getting most of a recommended daily dose.  Again I never will get around to considering how it works with toothpaste as adding exposure to fluoride here, or other potential sources.


For some people this may not be enough background yet, and I'd recommend checking on this reference:  Fluoride in Drinking Water: A Scientific Review of EPA’s Standards.  The short version:  they don't recommend continuous exposure to water sources at 4 mg/liter, even though they state the risks would be higher for a limited number of more sensitive individuals at that level.  The paper clarifies that all cases of municipal water fluoride content higher than 1.2 mg/l are relating to natural levels being higher from local contaminants (it can occur as a natural mineral in rocks).

Interpreted further, if levels higher than 4 mg/liter are occurring naturally they are either recommending or requiring that local water authorities treat those sources to reduce the levels, as described in that EPA reference:


A very small proportion of the public water systems nationwide have exceeded the drinking water standard for fluoride. In these cases, the high level of fluoride is generally the result of natural background resulting from the geologic composition of local soils and bedrock. When routine monitoring indicates that fluoride levels are above the MCL of 4 mg/L, the public water system must take steps to reduce the amount of fluoride so that it is below that level. 


Drinking private well or spring water sources would be a different subject; people would only know about that exposure if their water had been tested.


Fluoride level in tea


Finally!  This reference looks reasonable, "Fluoride Content in Tea and Its Relationship With Tea Quality" (there's a paper listing here or a Scribd upload here).  To keep this moving let's go straight to a table of measured results:




So the point there wasn't discussing fluoride as a good or bad thing, hence that paper title, but we can apply the results towards that anyway.  Old leaves contain a lot more fluoride; who knew.  At first I was reading this related to hei cha / pu'er aging, and wondering how the tea was picking up so much extra fluoride later during storage, but of course they're referring to plucking standards instead, using smaller, younger leaves towards the end of the branch versus older leaves lower on the branches.


Before we get to considering how much fluoride you would end up drinking one might wonder how this compares with using buds instead, which is on this table:




To rough out some of what this means, we're seeing a lower range of fluoride around 100-300 mg/kg for the second and third leaves, even lower for a bud and first leaf.  Fourth through sixth leaf levels are still lower than many in that prior "old leaves" table, which ranged from 500 mg/kg up to 2000, with some grouping in the middle, around 1200 or so.


There's one more hurdle before we can do the simple multiplication step:  these tables are measuring fluoride in fresh leaves, and I've just referenced how much dry tea someone might drink.  We need to factor the water removed from fresh leaves during processing back out first.  That would be easy, if I already knew how the relative weights of fresh and processed leaves compare, and I have heard that before, but it didn't stick.

Per this reference (a River Tea vendor article) fresh tea leaves contain 70 to 80% water, and per this World of Tea blog article reference processed dry tea contains 2 to 3 % moisture.  At the risk of completely botching conversion dry tea might well weigh one fourth what initial fresh leaves do, if around 75% of the weight in water is removed.  I just asked someone who just made some batches of tea (in Spain, a long story) and she confirmed that dry tea weight versus fresh leaves can vary from one fourth to one third original weight, per her experience, which does vary based on different factors.


Editing note:  after checking with a tea producer in Wuyishan prepared teas can be reduced by 80-90% of original fresh leaf weight, which again depends on tea type and other factors.  The calculations following will be revised to multiply fresh tea levels by four to estimate prepared tea fluoride levels, a conservative estimate based on these inputs (a multiplier of 5 would be more correct based on only this input).


Or, alternatively, there is a table in that study measuring fluoride present in dry, prepared tea leaves. But what fun is that compared to sorting through how it should work out in theory (and the subject comes up again in evaluating these results against the range of what is possible in other tea types not listed):


already processed dry tea amounts measurements



An average amount doesn't exactly jump out, but 50 to 60 mg/kg might be a typical range for grade 1 or 2, with plenty of outliers on either side.  Fujian oolong runs a bit higher, 100 to 150 instead, with green teas lower for most, 60 or less, but one set of Zhejang green tea samples is in that range too.

It won't be so easy to translate this to what someone drinks in a day, or to account for absorption rates, etc.  To simplify things let's just assume someone will take in all the fluoride present.  Pointing out a range of how much tea someone can drink in a day is problematic, even a working estimate.  Let's say that 5 grams of dry tea can brew a good bit, a few cups, and people wouldn't tend to drink more than two or three times that, so a higher working range of 10 to 20 grams per day will do.

20 grams of dry tea a day is an awful lot, of course, but some people would get there, so we could use that as an upper limit, with 10 grams a day serving as a more reasonable substantial habit.


Fluroride levels in prepared (dry) tea calculation



Let's check on total fluoride based on some earlier estimates, using 20 grams per day at 60 mg. / kg:

20 grams dry tea X 60 mg fluoride / 1000 grams fresh leaves  =  1.2 milligrams


Not so bad, especially given that was drinking a lot of tea.  Some finished tea samples double that measured amoung, up to 2.4 mg / day instead, but even that level seems fine, unless someone was already taking a lot in from treated water.

The highest levels on that finished tea chart leveled off around 200 mg/kg (with one outlier at 252), which would bring daily fluoride consumption based on 20 grams of tea per day to 4 milligrams instead.  That's still a very safe amount, unless someone is taking in that much fluoride from water sources as well, and that's still within the WebMD guideline range (8 mg versus a 10 mg recommended limit).


Calculations related to higher fresh tea levels:



Lets check out how that works out with the fresh leaves calculation, to see where this stands related to leaf inputs.


Down at the 100 mg/ kg fluoride to fresh tea leaves level (see the last round of charts), if someone consumed 20 grams of dry leaves per day it would work out like this:


20 grams dry tea X 4 gm fresh leaves / 1 gm of dry leaf  X 100 mg. fluoride / 1000 gm fresh leaves


Equals 8 milligrams of fluoride, consumed per day.  That person in the example was drinking a lot of tea (20 grams); drinking 10 mg. of tea per day resulting in 4 mg. / day might be more sensible.

The reason that number jumped was because of the assumed much higher level of fluoride in a tea prepared from fresh leaves.  Here I used a multiple of 4 (dry tea weighing one third the fresh leaf weight) but note that none of the prepared leaves amounts on the third table actually measured at 400 mg fluoride / kgm of prepared (dry) leaves; all were below that, with only one sample measured around 250 and three around 200.


It seems that only younger leaves are being used to make teas, related to data in the fresh leaf samples and these measured finished tea results.  It's conceivable that fluoride could have been removed somehow in processing (eg. there was more in stems measured in the other set, which were sorted out), but it seems like the plucking standard--using younger leaves--is the likely explanation for the difference.


That last calculation was still on the low side related to the oldest leaf fluoride measurements though.  For "old leaves" 1000 mg / kg of fresh leaves was still moderate, so based on the same calculation as the last one that bumps daily fluoride intake to 40 to 80 mg / day, way over safe limits.

It is quite conceivable that a different tea type than those tested could contain much higher amounts of fluoride.


Research paper on fluoride measured in prepared Indian tea



It would be nice to review a paper testing amounts of fluoride measured in some prepared teas, to make sure these conclusions didn't miss something, and one did turn up:

Estimation of fluoride concentration in tea infusions, prepared from different forms of tea, commercially available in Mathura city


The numbers cited in that study are a great match for the others reviewed, including this clear summary of where it all stands, and what I might have missed:


It is important to note that the availability of F− for consumption is not only from drinking water but also from other sources such as diet, dairy products like milk, fruits and vegetables, beverages like tea and coffee, etc., This means that the beneficial or detrimental effect of F− will depend upon the total consumption from all sources taken together.

This table summarizes their test findings:




Nothing too scary in that, but it would be possible for someone to drink more than a liter of tea in a day.  In the earlier estimates someone drinking liquid prepared from over 10 mg of dry tea almost certainly would be.  One part of the conclusions states:  Indian population usually consumes on an average 150-200 ml of tea/day; not much compared to what I have with breakfast.

They are concluding this level of fluoride is probably generally more positive than negative, typically a beneficial supplement rather than a risk.  One fifth of this highest measured level (relating to 200 ml) is still around .76 mg of fluoride, not so much, and drinking a whole liter is far from a health risk given the entire testing range is still under 4 grams for that.


Conclusions:  summary of findings


That seems like a complete enough review based only on those inputs.  Here is a summary of it all.


Range of fluoride level (recommended versus at risk):

-recommended intake, 3 to 4 mg/day (perhaps not universally accepted; that's a "Web MD" reference)

-at-risk range, as low as 4 mg/day, maybe as high as 10 mg/day for a standard limit; individual tolerance and reactions vary (with a limit really relating to amount per body weight, not a simple measurement)



Range of fluoride level input already in municipal drinking water:

-2.4 milligrams per day (based on highest range of typical level, at 2 liters consumption per day)

-much higher in some locations if local contaminant sources (amount present as a natural mineral) is not treated and removed.


Some people could already be taking in too much fluoride from water, but only in cases where it's a natural contaminant, since the added municipal water level limit of 1.2 mg/liter is relatively safe.

The amount used as a universal control limit for municipal water could be clearer; the past control limit of 4 mg/liter may no longer be accepted as standard guidance.


Range of fluoride level in tea:


-an Indian research study found measured prepared tea levels in the range of 1 to 3.8 mg/liter of prepared tea.  That study assumes people would typically drink well under a liter per day (true for most people, I guess).

-working from measured dry tea levels is not a simple calculation, but perhaps 1.2 - 2.4 mg/day could be ingested based on consuming 20 mg dry tea per day per one study findings (a lot of tea).  At the highest prepared tea recorded levels in one cited study that could relate to 4 mg fluoride intake per day.   

-that total could be much higher if old leaves are used as tea source (fifth leaf on branch average is five times as high; older leaves content can be 10 to 20 times that high)

-potential consumption could be 40 to 80 mg/day for consuming tea sourced only from old leaves (based on that research data and some assumptions, and 10 - 20 mg of dry tea consumed per day).  That's a lot higher than what is considered long-term safe levels, but none of the measured finished tea levels were anywhere near that.  It seems possible that causes that offset produced tea fluoride levels versus measured fresh tea leaf amounts for the oldest leaves weren't taken into account.


Final conclusions


So most people are probably safe enough, related to tea being a primary source of fluoride.  An easy option to reduce fluoride intake is to drink better tea, those sourced from buds and smaller leaves versus older leaves.

It seems possible that something was missed in this related to very low grades of tea, perhaps for tea produced in other countries, or versions of hei cha made more typically from older leaves instead.  I'm reminded of a comment by a tea vendor from Turkey saying instead of the plucking standard there relating to use of a leaf and two buds it was more like using the first foot or two of a branch.  Maybe that was just a joke.

Comments in that Steepster thread discussion and in this one in Tea Chat both mentioned some types of hei cha are known for containing relatively high fluoride levels, but I found no references to put numbers to that.  If those do relate to the "old leaves" measured quantities in one study cited there could be some risk related to regular consumption.  Otherwise the risk of consuming fluoride in tea seems limited, even for drinking lots of tea, unless other sources already elevate the amounts consumed.

The normal municipal water level should be safe (1.2 mg/l or less), even combined with substantial tea consumption.  Ironically the most risk might come from using spring or well water sources where natural levels of fluoride aren't monitored and happen to be high (shown in that area risk map).


No comments:

Post a Comment