The gut and excretion: why the final decision about elimination is made in the gut
The liver does the work. It attaches a water-soluble tag to fat-loving substances and hands them over to the gut via the bile. What happens next is decided down there. And down there sits an enzyme that can reverse the whole calculation.
People usually talk about detoxification as though the liver were the exit. It is not. It is the mailroom. The exit is the gut.
May I ask you an uncomfortable question? How often in the past few weeks have you thought about whether your gut actually empties completely?
I am not asking because I want to sell you a cure programme. I am asking because in conversations I keep seeing the same pattern. People invest a great deal of attention in what goes into the body. And very little in the question of what actually comes out again.
That is not a reproach. It has to do with the way this topic is discussed. Detoxification is told as a liver story, with bitter herbs, milk thistle and cures. The biochemical part of that is correct too. It is just that the story ends where, physiologically, it is only starting to get interesting.
What this article covers
- What phase I and phase II in the liver actually do
- Why glucuronidation is an address label and not destruction
- How the bile hands the substances over to the gut
- Beta-glucuronidase: the bacterial enzyme that can remove the label
- What human data from drug therapy show about this
- Estrobolome: estrogen in the enterohepatic circulation
- Bilirubin, the textbook example from every children's hospital
- Transit time, dietary fibre and bile flow as the three real levers
- Binders: what is documented, what is not, and where the risks lie
What actually happens in the liver
In everyday language the word detoxification has become an emotional term. In physiology the same process is called biotransformation and excretion. That sounds more sober. But it is more precise, and it immediately makes clear why the gut is at the centre.
The body's basic problem is a solubility problem. Many substances that you take in or produce yourself are fat-loving. They dissolve well in membranes and in fatty tissue, but poorly in water. Excretion, however, happens through water: through urine and through the watery bile. So a fat-loving substance cannot simply leave. It would first have to be rebuilt.
That is exactly what the liver does, in two consecutive steps.
Phase I: a handle is created
Enzymes of the cytochrome P450 family attach a reactive group to the molecule, usually by oxidation. This does not make the substance harmless. It makes it accessible. Some substances are even converted into a more reactive form in this step, which toxicology describes as bioactivation.
ActivationPhase II: the address label is applied
A water-soluble residue is coupled to this handle. The most common one is glucuronic acid, hence the name glucuronidation. Other routes are sulfation, coupling to glutathione or to amino acids. The result is a conjugate: the same substance, now with a water-soluble tag and markedly reduced biological activity.
ConjugationPhase III: transport out of the cell
Conjugates do not walk out by themselves. Transport proteins in the cell membrane actively pump them away. Smaller conjugates preferentially go via the kidney into the urine. Larger and more fat-loving parent molecules preferentially go via the bile.
ExportThe handover: bile into the duodenum
Whatever goes via the bile ends up in the small intestine. From here on the liver has done its part. The substance is in the gut lumen, so formally already outside the interior of the body. Whether it stays there is decided now.
HandoverThe fork in the road inside the gut
Two possibilities. Either the conjugate travels on with the stool and leaves the body. Or a bacterial enzyme detaches the tag again, the substance becomes fat-loving once more, and it can pass through the intestinal wall back into the blood. From there it goes via the portal vein back to the liver.
DecisionRowland, Miners and Mackenzie summarised the state of knowledge on the UDP-glucuronosyltransferases, UGT for short, in the International Journal of Biochemistry and Cell Biology in 2013.
What is described: humans have 22 of these proteins, divided into five families and six subfamilies. The subgroups UGT1A and UGT2B in particular terminate the biological effect of non-polar drugs from all therapeutic classes and improve their excretion. The liver carries the largest amount and variety of these enzymes, but they are also active outside the liver to varying degrees.
What this means for you: glucuronidation is not a side route. It is one of the body's central exits for everything that is fat-loving. And it produces a molecule that still exists. It is only labelled.
Rowland A, Miners JO, Mackenzie PI. Int J Biochem Cell Biol. 2013;45(6):1121-1132. DOI: 10.1016/j.biocel.2013.02.019Many people picture detoxification as an incinerator. Something goes in, something is destroyed, done. That is not how the body works. It destroys almost nothing. It packs, labels and transports. The molecule that was conjugated in phase II still exists afterwards, with a tag that makes it transportable.
That changes the decisive question. It is no longer: how hard is my liver working? It is: what happens to what my liver has already packed and handed over to the gut. And now you know why a liver-centred view of detoxification ends halfway along the route.
The bile is the handover, not the exit
Bile has a reputation as a digestive juice for fat. That is true, but it is only one of its jobs. At the same time it is the liver's outward channel. What the liver has pulled out of the blood and conjugated leaves it to a considerable extent along this route.
And here it is worth looking closely. The body built in the enterohepatic circulation not as a glitch but as a principle. This is seen most clearly with the bile acids themselves.
Alan Hofmann summarised the enterohepatic circulation of bile acids in mammals in Frontiers in Bioscience in 2009.
What is described: primary bile acids are formed from cholesterol in the liver cell, secondary ones arise in the lower part of the gut through bacterial conversion. Recovery in the gut shapes the composition of what later appears in the bile again. Because the liver retrieves them very efficiently, blood levels stay low, and practically nothing is excreted via the kidney.
What this means for you: reabsorption from the gut is an entirely normal, intended capability of the body. It is not limited to bile acids. Anything that looks chemically similar enough can take the same route.
Hofmann AF. Front Biosci (Landmark Ed). 2009;14(7):2584-2598. DOI: 10.2741/3399Bile acids are themselves signalling molecules as well. A comprehensive review in Physiological Reviews described them in 2018 as active regulators of the intestinal lining, with their own receptors on the epithelial cells, with influence on transport processes and on barrier function. They are at once an outward channel, a digestive agent and a messenger. That is typical of the body, and it is why a sluggish bile flow rarely concerns only one thing.
Without transport, even the best packaging just sits there. The liver makes the parcel. The gut takes it out. Or not.
Beta-glucuronidase: the enzyme that peels off the label
Now comes the part I almost never find in guide articles about detoxification, although it is the core of the matter.
Beta-glucuronidase is an enzyme that separates exactly one bond: the one between a substance and the glucuronic acid residue attached to it. It undoes the work of phase II. The human body has such a variant of its own, yet by far the larger part of the activity in the gut lumen comes from bacteria.
How large this bacterial apparatus is was shown systematically for the first time by a mapping study.
A group around Pollet at the University of North Carolina analysed the Human Microbiome Project data for the gastrointestinal tract in Structure in 2017 and searched specifically for beta-glucuronidase genes.
What was observed: 3,013 such proteins in total, among them 279 unique variants that could be sorted into six structural classes. 112 of these had been unknown until then. The authors describe marked variability between the 139 people studied. In laboratory experiments with the most common representatives there were also functional differences, among others in the handling of smaller glucuronides compared with larger carbohydrates.
What this means for you: beta-glucuronidase is not a switch you set to high or low. It is a whole enzyme family with different preferences. And how much of it works in your gut differs from person to person.
Pollet RM, D'Agostino EH, Walton WG et al. Structure. 2017;25(7):967-977.e5. DOI: 10.1016/j.str.2017.05.003That this enzyme is not merely a biochemical curiosity first became properly clear in cancer medicine. And it did so by an unpleasant detour.
Wallace and colleagues reported in Science in 2010 on the dose-limiting side effect of the colorectal cancer drug CPT-11, namely severe diarrhoea. The explanation lay not in the drug itself but in the gut: symbiotic bacterial beta-glucuronidases reactivate the excreted active substance there.
What was observed: in a high-throughput screen, inhibitors were found that blocked the bacterial enzyme while leaving the corresponding human enzyme untouched. Crystal structures showed that this selectivity rests on a loop that only bacterial variants possess. The inhibitors did not kill the bacteria and did not damage human cells. Given orally, they protected mice from the toxic effect of the drug in the gut.
What this means for you: here it was cleanly shown for the first time that reactivation in the gut lumen is not an afterthought but a distinct process that can be influenced deliberately. The evidence comes from cell experiments and the mouse model, not from a study in patients.
Wallace BD, Wang H, Lane KT et al. Science. 2010;330(6005):831-835. DOI: 10.1126/science.1191175The same principle turned up shortly afterwards with an entirely different group of medicines. A group around Saitta showed in Xenobiotica in 2014 that an inhibitor of bacterial beta-glucuronidase could also protect mice from small intestinal damage caused by indomethacin, ketoprofen and diclofenac. The timing finding was remarkable: complete protection when the inhibitor was given before the painkiller, partial protection still three hours afterwards. The authors explain this by the fact that the window for protecting the mucosa is extended for substances with a pronounced enterohepatic circulation.
The circulation can also be observed in the direction of the liver. A study in rats in Hepatology found in 2018 that animals with a stronger liver reaction to the active substance tacrine showed a 3.3-fold higher systemic exposure, together with increased return of the uncoupled substance via the gut. Metagenomic analysis showed roughly 9 percent higher abundance of beta-glucuronidase genes in these animals. When the researchers added the bacterial enzyme from outside, or conversely altered the gut flora with antibiotics, the animals' sensitivity could be shifted in both directions.
The gut is not a passive pipe through which waste falls. It is a biochemically active place where decisions of the liver can be revised.
The core of this articleWhat of this is measurable in humans
Now the honest question you should be asking at this point: is all of this mouse and test tube, or are there human data?
They exist. Almost all of them come from drug therapy, because blood levels can be measured precisely there. For classic environmental pollutants the human data are considerably thinner. That belongs to being truthful about this topic.
A group around Drevland in Oslo studied kidney transplant recipients taking the immunosuppressant mycophenolate in Microbiome in 2025. With this drug, blood levels vary greatly between people, and one main reason is the differently pronounced enterohepatic circulation.
What was observed: the stool microbiomes of 21 transplant recipients were compared before and after transplantation with those of 15 healthy people. For each person it was determined how quickly their microbiome converts the excreted conjugate back into the active substance. This rate correlated clearly and positively with the actual extent of the enterohepatic circulation in the body. Certain gene variants of beta-glucuronidase associated with Faecalibacterium prausnitzii had a particularly strong influence.
What this means for you: this is the most direct human evidence so far. The composition of your microbiome can help determine how long a substance circulates in your body before it leaves.
Drevland OM, de Muinck EJ, Trosvik P et al. Microbiome. 2025;13(1):169. DOI: 10.1186/s40168-025-02142-6How individual this splitting capacity is was measured more closely by a Finnish study in Drug Metabolism and Disposition in 2025. Stool samples were examined from 12 healthy donors, three faecal transplantation donors and seven patients with recurrent Clostridioides difficile infection. Between people, the ability to detach the tag varied considerably, particularly for steroid conjugates. Within one person, by contrast, it remained fairly stable over time.
In the same work, the patients with recurrent Clostridioides difficile infection had a particularly low splitting capacity at the start. After the faecal transplant it rose. The authors therefore explicitly describe the removal of glucuronides as a function of a healthy gut microbiome.
That does not fit the story that beta-glucuronidase is the blanket enemy of excretion. A low value can also mean that the gut flora is damaged. I consider this the most important sentence of this whole section. Anyone who presses this mechanism into a simple good versus bad scheme shortens it so much that false conclusions follow.
Metabolism lens
Glucuronidation consumes glucuronic acid, which comes from carbohydrate metabolism. Coupling to glutathione consumes glutathione. Conjugation is not something the body gets for free. It depends on substrates and energy.
Immune lens
According to a review in Physiological Reviews, bile acids can act on the intestinal lining and its barrier function through their own receptors. A disturbed lining therefore also changes the conditions for uptake and re-uptake at the same time.
Nervous system lens
Intestinal passage is co-regulated by the enteric nervous system and appears to be closely coupled to stress, sleep and movement. Anyone who wants to influence transit time should therefore look at these three topics as well.
Hormonal lens
Estrogens run along exactly the same route of conjugation, biliary release and possible reactivation in the gut. That is not a footnote but a research field of its own, called the estrobolome.
Estrobolome: estrogen in circulation
If you have been thinking so far that this is a purely pharmacological topic, here is the point where it becomes personal.
Estrogens are conjugated in the liver, among other routes glucuronidated, and partly released into the gut via the bile. There, bacterial beta-glucuronidase can convert them back into the free, active form. The share of gut bacteria capable of this is called the estrobolome.
A group around Ervin tested systematically for the first time, in the Journal of Biological Chemistry in 2019, which of the beta-glucuronidases from the human gut can reactivate two particular estrogen conjugates, namely estrone-3-glucuronide and estradiol-17-glucuronide.
What was observed: of 35 enzymes studied, certain representatives from three structural classes were able to release the estrogens from their inactive conjugates. The reactivation could be inhibited both in the purified enzymes and in stool preparations from mice. A targeted inhibitor did not, however, prevent tumour development in an established breast cancer mouse model.
What this means for you: this is the first laboratory demonstration that enzymes from gut bacteria can release estrogens from their conjugates. Whether and how strongly that co-determines a person's estrogen balance is not answered by it. The leap from this mechanism to disease prevention was explicitly not achieved. The authors themselves describe the estrobolome as a multi-layered system with presumably many enzymes involved.
Ervin SM, Li H, Lim L et al. J Biol Chem. 2019;294(49):18586-18599. DOI: 10.1074/jbc.RA119.010950At the level of observation in humans there are matching findings. A cross-sectional study of 60 healthy postmenopausal women found in the Journal of Clinical Endocrinology and Metabolism in 2014 that the ratio of estrogen metabolites to the parent hormones in urine was related to the phylogenetic diversity of the stool microbiome, with a correlation coefficient of 0.35 at a p value of 0.01. Associations also appeared for the order Clostridiales and for the genus Bacteroides, in the opposite direction.
The obvious thought runs: less reactivation in the gut means less estrogen in the blood, and less estrogen is good. It is not that simple.
A review in Maturitas described the reverse case in 2017. In a dysbiosis with lower microbial diversity, reduced removal of the tag can lead to less circulating estrogen. After menopause, when estrogen is scarce anyway, that is no advantage. An exploratory randomised trial in peri- and postmenopausal women went exactly in this direction in 2024. In the group taking a probiotic formulation containing a strain with beta-glucuronidase activity, estrogen levels remained stable on average over twelve weeks, while they fell under placebo. Two limitations belong alongside this: the trial was explicitly designed as a first exploration, and it was carried out by employees of the companies manufacturing the bacterial strain. No recommendation can be derived from it yet, only the direction that the gut plays a part in the estrogen circulation.
So the gut does not simply hold estrogen back. It participates in its circulation, and this participation can act in both directions, depending on life stage and situation. And now you know why I am cautious when someone declares a single enzyme value to be the target.
Bilirubin: the classic every children's hospital knows
If you are wondering whether this whole circulation actually plays a role in clinical practice, then look at the neonatal ward.
Bilirubin arises from the breakdown of the red blood pigment. In the liver cell it is glucuronidated by the enzyme UGT1A1 and released via the bile. In the gut, beta-glucuronidase can split this conjugate again, and the free bilirubin can be taken back up. In newborns this route meets an enzyme system that is still immature and a gut flora that is still young. Part of the familiar neonatal jaundice arises from exactly that.
A group around Tsai in Taichung investigated in Medicine in 2022 whether a specifically selected bacterial strain can support phototherapy in neonatal jaundice. The basis was the idea that gut bacteria with high beta-glucuronidase activity can strengthen the enterohepatic circulation of bilirubin.
What was observed: 83 newborns with a bilirubin value of 15 milligrams per decilitre or above were randomised. The group receiving additional Bifidobacterium animalis subsp. lactis CP-9 showed a faster fall in bilirubin at minus 0.16 milligrams per decilitre per hour at a p value of 0.009, as well as a shorter total phototherapy duration of 44.82 hours at a p value of 0.011. The effect was clearest in full-term newborns.
What this means for you: the circulation is not a theoretical quantity. It is clinically measurable, and in a clearly defined situation it can be influenced. This is a single study in a very specific group, not a general recipe.
Tsai ML, Lin WY, Chen YT et al. Medicine (Baltimore). 2022;101(45):e31030. DOI: 10.1097/MD.0000000000031030In adults there is a counterpart on the liver side. In Gilbert's syndrome UGT1A1 works more slowly for genetic reasons. A study in Egyptian adults found a frequency of around 8 percent in the group examined and described anaesthesia, pregnancy, fasting phases beyond twelve hours, low-calorie weight loss plans, infections and heavy physical strain as typical triggers of visible yellowing. Gilbert's syndrome is regarded as benign, and it stays that way. The same work shows, however, that the episodes can noticeably affect wellbeing and that medical support makes sense, particularly around anaesthesia, pregnancy and weight loss phases. One more thing matters: when UGT1A1 works more slowly, the body can be slower to process certain drugs. Tell your doctor about it before a new therapy starts. It shows very vividly that phase II has a capacity which differs from person to person.
Transit time, dietary fibre and bile flow
From all of this follows a very concrete question. If the fork in the road lies in the gut, what then influences which route a conjugate takes?
Three quantities are mechanistically plausible and at least partly studied in humans: the time the gut contents need, the amount and kind of what is transported with them, and the bile flow itself.
The time
A Danish group around Roager determined colonic transit time with radio-opaque markers in Nature Microbiology in 2016 and compared it with the composition, diversity and metabolic output of the gut microbiome.
What was observed: a long passage time went along with high microbial diversity, but at the same time with a shift in metabolism away from carbohydrate fermentation and towards protein breakdown. This was measurable as higher concentrations of protein-derived, potentially unfavourable breakdown products in the urine. A shorter passage time, by contrast, correlated with metabolic products suggesting stronger renewal of the intestinal lining.
What this means for you: the authors draw a conclusion from this that matters for our topic. High microbial diversity does not automatically mean a healthy gut ecosystem. And transit time is such an influential quantity that it belongs in the thinking of any microbiome study.
Roager HM, Hansen LBS, Bahl MI et al. Nat Microbiol. 2016;1(9):16093. DOI: 10.1038/nmicrobiol.2016.93For excretion, a very plain mechanical argument also remains. The longer a conjugate lies in the gut, the more opportunity there is for it to be split there. That is not a measured figure, that is a physical consideration. I name it as such.
The dietary fibre
Here there are human data, and surprisingly old ones.
Goldin, Adlercreutz, Gorbach and colleagues studied ten vegetarian and ten omnivorous premenopausal women in the New England Journal of Medicine in 1982, at four appointments roughly four months apart. Estrogens were measured in blood, urine and stool.
What was observed: the vegetarian group took in less fat, namely 30 versus 40 percent of calories, and more fibre, namely 28 versus 12 grams per day. Stool weight was closely related to estrogen excretion via the stool in both groups, at a p value below 0.001. Blood values for estrone and estradiol were negatively linked to this excretion, at a p value of 0.005. In addition, bacterial beta-glucuronidase activity in the stool of the vegetarian women was lower.
What this means for you: this is not a randomised intervention trial but a comparison of two ways of eating. Still, it is a remarkably complete picture: diet, stool volume, enzyme activity, excretion and blood levels were all captured in the same study and pointed in the same direction.
Goldin BR, Adlercreutz H, Gorbach SL et al. N Engl J Med. 1982;307(25):1542-1547. DOI: 10.1056/NEJM198212163072502The same group summarised their findings in Preventive Medicine in 1987 and explicitly described the relationship as dietary influence on the enterohepatic circulation. Vegetarian women accordingly excreted three times the amount of estrogen via the stool and had 15 to 20 percent lower estrogen values in the blood. In a group of women who had immigrated from Asia with a very low-fat diet of 20 to 25 percent of calories, blood values were 30 percent lower.
The bile flow
The third point is named least often and is the simplest. If hardly any bile reaches the gut, hardly any of what the liver wanted to release along that route gets in either. Fat and protein in the duodenum are among the strongest stimuli for emptying the gallbladder, mediated by the hormone cholecystokinin. If very little fat arrives over a long period, less bile may flow into the gut as well. A high-quality plant oil or an algae-derived omega-3 oil with meals is the unspectacular way to take that into account.
If you have gallstones, this very stimulus can trigger colic, because a contracting gallbladder then presses against a stone. Gallstones are common, and many people do not know they have them. Cramping pain in the right upper abdomen after a fatty meal that radiates into the right shoulder belongs in a medical assessment, not answered with more oil. With severe pain plus fever, chills or yellowing, call the emergency number 112 or go straight to an emergency department.
And if a low-fat diet has been prescribed for you medically, for example after pancreatitis, with very high triglycerides or after gallbladder removal, then that prescription applies and not what you read here. Discuss any change with your doctor beforehand.
It is not the amount your liver conjugates that decides your excretion. It is whether it also leaves the gut.
Binders: what is documented and what is not
This is exactly the point where advertising claims and the data online diverge the most. That is why I look particularly closely here at what has been studied and what has not. If a substance can be reabsorbed in the gut, then the thought of binding it there before that happens is an obvious one. The thought is correct. It has also been studied. And it is very much worth looking closely at where it was studied.
Cohn and colleagues reported in the New England Journal of Medicine in 1978 on industrial workers who had been exposed to the insecticide chlordecone and showed signs of poisoning in several organ systems.
What was observed: in 22 of those affected the substance disappeared only slowly from the blood, with a half-life of 165 days, and from fatty tissue with around 125 days. The ratio of the routes was striking: output via the bile was ten to twenty times higher than what arrived in the stool. That pointed precisely to re-uptake in the gut. Colestyramine, also written cholestyramine, an anion exchange resin that binds the substance, increased excretion via the stool sevenfold. Over a five-month treatment the half-life fell to 80 days in the blood and to 64 days in fatty tissue.
What this means for you: this was the first clean demonstration that a binder in the gut can markedly speed up the excretion of a fat-loving foreign substance. On the explanation I am more careful than the 1978 paper suggests. The same research group found a year later that the substance still reached the gut when the bile was diverted to the outside. The route into the gut therefore appears not to run only via the bile, but partly directly across the intestinal wall. That does not change the observed benefit of the binder, but it shows how carefully one should handle the word proof. And it was a study in severely poisoned workers under medical supervision, not in healthy people in a wellness setting.
Cohn WJ, Boylan JJ, Blanke RV et al. N Engl J Med. 1978;298(5):243-248. DOI: 10.1056/NEJM197802022980504The follow-up work comes from the same research group. Boylan and colleagues described in Clinical Pharmacology and Therapeutics in 1979 that chlordecone still appeared in the stool when the bile was diverted to the outside, and that colestyramine increased excretion in that situation too. The authors concluded that there is a nonbiliary source, probably the gut wall itself. The observed benefit of the binder stands, the mechanistic explanation was qualified by its own originators.
Colestyramine is a prescription-only medicine. It is licensed for lowering raised blood lipids and for diarrhoea caused by bile acid loss, not for detoxification or elimination. Everything you read above comes from the treatment of a poisoning under medical supervision. Using it for elimination would be outside the licence, that is, off label. In off-label use the prescribing doctor carries the responsibility, and statutory insurers usually do not pay.
I deliberately give no dosages here. Known risks include reduced uptake of medicines taken at the same time and of certain fat-soluble vitamins, constipation, very rarely bowel obstruction, and a moderate rise in triglycerides. In bowel obstruction and in complete biliary obstruction colestyramine must not be given.
For activated charcoal the picture is more sober than the advertising suggests. The joint position statement of the American Academy of Clinical Toxicology and the European Association of Poisons Centres and Clinical Toxicologists from 1999, still the reference point for this approach today, states for multi-dose activated charcoal: numerous studies in animals and volunteers do show accelerated excretion, but there is no controlled study in poisoned patients demonstrating lower mortality or fewer complications. It is recommended explicitly only for life-threatening ingestion of carbamazepine, dapsone, phenobarbital, quinine or theophylline. Without an intact or protected airway it is contraindicated, and it must not be given in bowel obstruction. Giving laxatives alongside it is explicitly not recommended in the same statement. In small children it warns against this in addition, because laxatives there can derail fluid and electrolyte balance.
Binders do not distinguish between what you want to get rid of and what you want to keep. For bile acid sequestrants, studied in their actual role as lipid-lowering drugs, a review in the American Journal of Cardiology describes three kinds of unwanted effects: reduced uptake of medicines taken at the same time and in part of certain vitamins, a change in gut contents with constipation and very rarely bowel obstruction, and a moderate rise in blood triglycerides. For use as a binder for elimination there is no safety review of its own, which is why I transfer these points, and why they are more a lower bound than a complete list.
In practice this means: time spacing from all medicines and from nutrient supplements, plenty of fluid, no permanent use without a reason, and a medical conversation beforehand if you take medicines regularly. This applies particularly to thyroid hormones, anticoagulants, immunosuppressants and hormonal contraception.
And the point that decides everything else: a binder without regular bowel movements is pointless to risky. It then binds something that stays in the gut anyway.
Between what stands above and what is sold online under the heading of elimination lies a large gap. What is documented: the enterohepatic circulation exists, bacterial beta-glucuronidase can split conjugates again, and in a severe poisoning with a lipophilic substance a binder can speed up excretion.
What is not documented: that daily intake of binders in healthy people with everyday environmental exposure measurably lowers the body burden or improves complaints. To my knowledge such studies do not exist in robust form. Anyone claiming otherwise should be able to name the source.
What follows from this for your everyday life
If I boil all of this down to three levers that are, from my point of view, mechanistically grounded and workable in everyday life, these are they. A manageable risk profile does not mean risk-free, so each point says what to watch out for.
1. Complete emptying, every day
This is the least spectacular and most important point of the whole article. For everything the liver releases via the bile, the stool is the main way out. What is split and taken up again in the gut can later go via the kidney, but for that the liver has to package it again. Every lap of the circulation therefore costs work. If you have lived with constipation for years and have filed it away as normal, this is the place where I would ask you to look more closely.
2. Plant variety and dietary fibre
In the available human data on estrogens, stool weight was very closely related to excretion via the stool. Not a single supplement, but the amount and variety of plant components in the daily diet. It matters to increase the amount slowly and to drink enough, otherwise bloating and a feeling of fullness can occur. If you have a chronic inflammatory bowel disease, a known narrowing in the gut or previous abdominal surgery, do not increase fibre on your own, discuss it medically first. And if you take thyroid hormones, keep time between the tablet and large amounts of fibre, because uptake can otherwise fluctuate.
3. Fat with meals and brassica vegetables
Fat stimulates bile flow, and without bile there is no biliary release. If you have gallstones or have been prescribed a low-fat diet, the caution box above applies to you and not this point. Brassica vegetables are worth a point of their own, with a clear classification. A review from cancer research describes, for phenethyl isothiocyanate, a substance formed from the mustard oil glycosides during chewing and found above all in watercress, two things: an inhibition of certain activating enzymes and a stimulation of phase II enzymes, among them glucuronosyltransferase. These enzyme data come mainly from animal models, and no numerical promise for humans can be derived from them. And because the same mechanism runs via cytochrome P450, a very large amount of brassica vegetables could in theory also affect the metabolism of medicines. Anyone taking warfarin-type anticoagulants should also avoid abrupt changes in the amount of brassica, because it is rich in vitamin K. As a normal part of a varied diet it remains a good idea nonetheless.
And what I would not recommend
A cure programme laid out for a weekend or ten days. The circulation this article is about runs on every single day of your life. It has no interest in calendar sections. And a binder without functioning bowel movements is the wrong order of operations.
When this topic belongs in a medical assessment
- Persistently changed bowel habits over several weeks, in either direction
- Blood in the stool or black, tarry stool, regardless of the amount
- Newly appeared yellowing of the skin or the whites of the eyes
- Unexplained weight loss without a change in diet or movement
- Severe or persistent abdominal pain, particularly with fever. Severe abdominal pain with fever, chills or yellowing is an emergency, then call 112
- Cramping pain in the right upper abdomen after a fatty meal that radiates into the right shoulder, as a possible sign of gallstones
- Regular use of medicines, before you add binders or high amounts of fibre
- Pregnancy and breastfeeding, because separate rules apply here and data are often missing
What holds me to this topic is, in the end, not the biochemistry. It is the change of perspective behind it. We talk about detoxification as though it were a project you start. It is not one. It is running while you read this sentence, in every liver cell and in every metre of your gut.
And the place where you actually get a say is not the liver. It is the exit. A body that lets go is not an optimised body. It is a body you are not constantly giving extra work to. That is a different kind of freedom from the one cure programmes promise, and from my point of view a more durable one. And now you know why.
Where this topic also plays in
Excretion does not stand on its own. The topic touches the liver, the intestinal lining, the microbiome and hormone balance all at once. You will find further articles on this gathered in the following sections.
The gut and excretion
Enterohepatic circulation and beta-glucuronidase
this articleGuide Detoxification
All articles on biotransformation, exposures and excretion
Guide Gut Health
Mucosa, microbiome and digestion
All guides
The complete ViveCura Blog at a glance
Frequent questions about the gut, excretion and detoxification
What is the enterohepatic circulation, put simply?
The liver makes fat-loving substances water-soluble by attaching a small tag made of glucuronic acid. It releases a large share of this conjugate into the small intestine via the bile.
From there the conjugate can take two routes: out with the stool, or back again. It goes back when gut bacteria carrying the enzyme beta-glucuronidase detach the tag again. The substance is then fat-loving once more and can pass through the intestinal wall back into the blood and via the portal vein to the liver.
This back and forth between liver, bile, gut and blood is called the enterohepatic circulation. It is not a malfunction but normal physiology. With bile acids the body even uses it deliberately as a savings programme, as a review by Hofmann describes.
What is beta-glucuronidase and why does it matter?
Beta-glucuronidase is an enzyme that separates exactly one bond, namely the one between a substance and the glucuronic acid residue attached to it. A large part of this activity in the gut lumen comes from bacteria.
A mapping study of the human gut microbiome found 3,013 such proteins in total, among them 279 different variants across six structural classes, and it described marked differences between the 139 people studied. In laboratory experiments, too, the most common representatives differed functionally from one another.
For excretion this is decisive, because only the removal of the tag turns a conjugate that was ready to leave back into a substance that can be reabsorbed. One point of context: a lot or a little beta-glucuronidase is not good or bad across the board, it depends on the particular substance and the situation.
Is there human evidence for this, or only laboratory data?
Both, with a clear emphasis. The mechanism is very well documented in the laboratory and in animal models, first in a paper in Science on the colorectal cancer drug CPT-11 and later with anti-inflammatory painkillers in the mouse model.
In humans, the clearest data come from drug therapy. A study in kidney transplant recipients compared 21 patients with 15 healthy people and found a clear association between the ability of the stool microbiome to split a drug conjugate again and the actual extent of the enterohepatic circulation in the body. A second study in healthy stool donors showed that this splitting capacity varies considerably between people, while remaining fairly stable within one person over time.
For classic environmental toxins the human data are considerably thinner than for medicines. With this topic that belongs to being honest.
What does the gut have to do with estrogen?
Estrogens are conjugated in the liver and partly released into the gut via the bile. There, bacterial beta-glucuronidases can convert them back into the free form, so that they can be taken up again. The group of gut bacteria with this ability is called the estrobolome.
In the laboratory, 35 human gut enzymes were tested for their ability to reactivate estrone-3-glucuronide and estradiol-17-glucuronide, and some classes were able to do so. At the observational level, a study of 60 postmenopausal women found associations between the diversity of the stool microbiome and the pattern of estrogen metabolites in the urine.
This is not a one-way street towards too much estrogen. After menopause the same reactivation may even help to support estrogen levels. For practice this mainly means: the gut is a participant in hormone balance, not a side stage.
Why does constipation matter for excretion?
Because the time the gut contents spend in the colon changes the chemistry there. In a Danish study using radio-opaque markers, a long colonic transit time went along with a shift in metabolism, away from carbohydrate fermentation and towards protein breakdown, measurable as higher levels of protein-derived breakdown products in the urine.
On purely mechanical grounds there is more: the longer a conjugate lies in the gut, the more opportunity there is for it to be split there again. That is a physical consideration, not a measured figure, and I name it as such.
This is not automatic. The same work also showed that a long transit time went along with greater microbial diversity, and that diversity alone is not yet a seal of quality for a healthy ecosystem.
Does dietary fibre really do anything for excretion?
For estrogens there are surprisingly old and surprisingly clean data. In a 1982 study in the New England Journal of Medicine, ten vegetarian and ten omnivorous women were compared at four appointments over about a year. The vegetarian group took in an average of 28 grams of fibre per day, the other group 12 grams.
Stool weight was closely related to estrogen excretion via the stool, and blood estrogen levels were inversely linked to that excretion. In addition, bacterial beta-glucuronidase activity in the stool of the vegetarian women was lower. A follow-up paper by the same group described a threefold higher excretion via the stool and 15 to 20 percent lower estrogen values in the blood.
This is not a causal chain from a randomised trial, but it is a very consistent pattern. For everyday environmental pollutants no numerical promise can be derived from it.
Do binders such as activated charcoal or cholestyramine make sense?
They can make sense in clearly defined situations, and that is exactly where they have been studied. Colestyramine, also written cholestyramine, is a prescription-only medicine. It is licensed for lowering raised blood lipids and for diarrhoea caused by bile acid loss, not for detoxification or elimination.
In poisoning with the insecticide chlordecone it increased excretion via the stool sevenfold in a controlled study of 22 exposed workers under medical supervision and shortened the blood half-life from 165 to 80 days. Using it for elimination would be outside the licence, that is, off label, and the prescribing doctor carries the responsibility for that. I deliberately give no dosages. Risks include reduced uptake of medicines taken at the same time and of certain vitamins, constipation, very rarely bowel obstruction and a moderate rise in triglycerides. In bowel obstruction and in complete biliary obstruction it must not be given.
For multi-dose activated charcoal, the joint position statement of two toxicology societies foresees use only in life-threatening ingestion of a few substances, namely carbamazepine, dapsone, phenobarbital, quinine and theophylline. It states explicitly that there is no controlled evidence of benefit in poisoned patients.
For long-term use in healthy people with everyday environmental exposure, this evidence does not exist to my knowledge. That is an important difference from what is often promised online.
What are the risks of binders?
Binders do not only bind what you want to get rid of. For bile acid sequestrants, a review describes three kinds of unwanted effects: reduced uptake of medicines taken at the same time and in part of certain vitamins, a change in gut contents with constipation and very rarely bowel obstruction, and a moderate rise in blood triglycerides.
For activated charcoal there is more: it is contraindicated without an intact or protected airway and must not be given in bowel obstruction. Giving laxatives alongside it is explicitly not recommended in the position statement. In small children it warns against this in addition, because laxatives there can derail fluid and electrolyte balance.
That is why time spacing from medicines and nutrients, sufficient fluid and medical supervision are not a formality, they are the actual point. Thyroid hormones, anticoagulants, immunosuppressants and hormonal contraception need particularly careful discussion.
What about bilirubin, is that the same route?
Yes, bilirubin is the textbook example of exactly this circulation. It is glucuronidated in the liver by the enzyme UGT1A1 and released via the bile, and in the gut beta-glucuronidase can split the conjugate again.
In Gilbert's syndrome UGT1A1 works more slowly for genetic reasons, which can lead to recurring yellowing. A study in Egyptian adults described anaesthesia, pregnancy, fasting phases beyond twelve hours, low-calorie weight loss plans, infections and heavy physical strain as typical triggers. The syndrome is regarded as benign, but the same work describes that the episodes can noticeably affect wellbeing. And when UGT1A1 works more slowly, the body can be slower to process certain drugs. Tell your doctor about it before a new therapy starts.
One important frame first: jaundice in a newborn always belongs in paediatric observation and treatment, never in self-treatment. Bilirubin levels that are too high can harm the infant brain, which is why measurements in hospital are close-meshed.
In neonatal medicine the circulation has also been studied therapeutically. In a single randomised trial of 83 newborns with jaundice, bilirubin fell faster with the additional administration of a particular Bifidobacterium strain and phototherapy was shorter than with phototherapy alone. The effect was clearest in the subgroup of full-term infants. The trial was carried out together with the manufacturer of the bacterial strain, and that belongs in the picture too. It is a signal from a clearly bounded hospital setting, not an agent for home use.
Does that mean I should lower my beta-glucuronidase?
No, at least not as a blanket goal. First, beta-glucuronidase is not a single enzyme but a whole family with different preferences for different substrates.
Second, reactivation in the gut is not consistently undesirable. After menopause it may contribute to maintaining estrogen levels. And in a study of people after recurrent Clostridioides difficile infection the splitting capacity was low at first and rose after a faecal transplant, that is, in the course of a recovery. The authors therefore describe this ability as a feature of a healthy microbiome.
Third, research itself has shown that the simple calculation does not always work out: a targeted inhibitor did not prevent tumour development in a breast cancer mouse model. More useful than a single enzyme value are therefore the questions of transit time, diet, bile flow and the composition of the microbiome.
What can I actually do without sliding into detox marketing?
Three things have, from my point of view, the best basis and a manageable risk profile. Manageable does not mean risk-free. First, regular and complete bowel emptying. For everything the liver releases via the bile, the stool is the main way out.
Second, a diet with plenty of plant variety and enough fibre, because stool volume and stool weight were closely related to excretion via the stool in the available human data. Increase fibre slowly and drink enough. With a chronic inflammatory bowel disease, a known narrowing in the gut or previous abdominal surgery, discuss it medically first, and with thyroid hormones keep time between the tablet and the fibre.
Third, fat with meals, because fat and protein in the duodenum are among the strongest stimuli for emptying the gallbladder, mediated by the hormone cholecystokinin. A high-quality plant oil or an algae-derived omega-3 oil serves this purpose. One caveat: with gallstones this very stimulus can trigger colic, and a medically prescribed low-fat diet continues to apply. Clarify both medically beforehand.
Brassica vegetables deserve a point of their own, with a classification. A review from cancer research describes, for phenethyl isothiocyanate, a substance formed from the mustard oil glycosides during chewing, an inhibition of certain activating enzymes and a stimulation of phase II enzymes. These data come mainly from animal models. Anyone taking warfarin-type anticoagulants should avoid abrupt changes in the amount of brassica because of its vitamin K content. What I would not recommend is a ten-day cure programme. The circulation runs daily, it has no interest in calendar sections.
When should I have this checked by a doctor?
Whenever symptoms could lie behind it that need a diagnosis. Persistently changed bowel habits, unexplained weight loss, blood in the stool, newly appeared yellowing of skin or eyes, severe or persistent abdominal pain and fever belong in a medical examination and not under treatment with a binder. Cramping pain in the right upper abdomen after a fatty meal can point to gallstones. With severe abdominal pain plus fever, chills or yellowing, call the emergency number 112.
A conversation is also part of it if you take medicines regularly. Binders and strong changes in intestinal passage can affect the uptake of active substances. This applies particularly to thyroid hormones, anticoagulants, immunosuppressants and hormonal contraception.
In pregnancy and breastfeeding the rule is in any case: no self-experiments with binders or high doses of supplements, but discuss it beforehand. For many of these agents, robust safety data are lacking in this stage of life.
Scientific sources
All studies named in the text were checked via PubMed for existence, authorship, journal and year, and the statements were matched against the respective abstract. Where I use only one aspect of a paper, I say so. The study type is given in square brackets. That classification is my own and can be debated. Three papers are marked as background reading, they are not cited individually in the text.
- Rowland A, Miners JO, Mackenzie PI. The UDP-glucuronosyltransferases: their role in drug metabolism and detoxification. Int J Biochem Cell Biol. 2013;45(6):1121-1132. DOI: 10.1016/j.biocel.2013.02.019 [Review, enzymology and pharmacogenetics]
- Hofmann AF. The enterohepatic circulation of bile acids in mammals: form and functions. Front Biosci (Landmark Ed). 2009;14(7):2584-2598. DOI: 10.2741/3399 [Review, pathophysiology]
- Hegyi P, Maléth J, Walters JR, Hofmann AF, Keely SJ. Guts and Gall: Bile Acids in Regulation of Intestinal Epithelial Function in Health and Disease. Physiol Rev. 2018;98(4):1983-2023. DOI: 10.1152/physrev.00054.2017 [Narrative review, physiology]
- Pollet RM, D'Agostino EH, Walton WG et al. An Atlas of β-Glucuronidases in the Human Intestinal Microbiome. Structure. 2017;25(7):967-977.e5. DOI: 10.1016/j.str.2017.05.003 [In vitro and metagenomics, 139 people, 3,013 enzymes]
- Wallace BD, Wang H, Lane KT et al. Alleviating cancer drug toxicity by inhibiting a bacterial enzyme. Science. 2010;330(6005):831-835. DOI: 10.1126/science.1191175 [In vitro and in vivo, mouse, crystal structure analysis]
- Saitta KS, Zhang C, Lee KK, Fujimoto K, Redinbo MR, Boelsterli UA. Bacterial β-glucuronidase inhibition protects mice against enteropathy induced by indomethacin, ketoprofen or diclofenac: mode of action and pharmacokinetics. Xenobiotica. 2014;44(1):28-35. DOI: 10.3109/00498254.2013.811314 [In vivo, mouse, pharmacokinetics]
- Yip LY, Aw CC, Lee SH et al. The liver-gut microbiota axis modulates hepatotoxicity of tacrine in the rat. Hepatology. 2018;67(1):282-295. DOI: 10.1002/hep.29327 [In vivo, rat, metagenomics and pharmacokinetics]
- Drevland OM, de Muinck EJ, Trosvik P et al. Microbiome-derived reactivation of mycophenolate explains variations in enterohepatic recirculation in kidney transplant recipients. Microbiome. 2025;13(1):169. DOI: 10.1186/s40168-025-02142-6 [Human cohort study, n=21 plus n=15, with in vitro data]
- Uoti A, Neulasalmi O, Hiippala K et al. Characterization of fecal deglucuronidation activity in healthy subjects and in patients treated with fecal microbiota transplantation. Drug Metab Dispos. 2025;53(12):100205. DOI: 10.1016/j.dmd.2025.100205 [Human observational study, n=22, ex vivo]
- Ervin SM, Li H, Lim L et al. Gut microbial β-glucuronidases reactivate estrogens as components of the estrobolome that reactivate estrogens. J Biol Chem. 2019;294(49):18586-18599. DOI: 10.1074/jbc.RA119.010950 [In vitro, 35 human gut enzymes, plus in vivo, mouse]
- Baker JM, Al-Nakkash L, Herbst-Kralovetz MM. Estrogen-gut microbiome axis: Physiological and clinical implications. Maturitas. 2017;103:45-53. DOI: 10.1016/j.maturitas.2017.06.025 [Review]
- Fuhrman BJ, Feigelson HS, Flores R et al. Associations of the fecal microbiome with urinary estrogens and estrogen metabolites in postmenopausal women. J Clin Endocrinol Metab. 2014;99(12):4632-4640. DOI: 10.1210/jc.2014-2222 [Human cross-sectional study, n=60]
- Honda S, Tominaga Y, Espadaler-Mazo J et al. Supplementation with a Probiotic Formula Having β-Glucuronidase Activity Modulates Serum Estrogen Levels in Healthy Peri- and Postmenopausal Women. J Med Food. 2024;27(8):720-727. DOI: 10.1089/jmf.2023.k.0320 [RCT, exploratory, double-blind, placebo-controlled, 12 weeks]
- Martínez-Nortes ME, Carrascosa-Romero C, Ávila-Gálvez MÁ, Espín JC. Impact of long-term medication on estrobolome-associated β-glucuronidase and sulfatase activities: Implications for estrogen homeostasis in postmenopausal women. Maturitas. 2026;206:108830. DOI: 10.1016/j.maturitas.2026.108830 [Review, narrative. Background reading, not cited individually in the text]
- Tsai ML, Lin WY, Chen YT et al. Adjuvant probiotic Bifidobacterium animalis subsp. lactis CP-9 improve phototherapeutic treatment outcomes in neonatal jaundice among full-term newborns: A randomized double-blind clinical study. Medicine (Baltimore). 2022;101(45):e31030. DOI: 10.1097/MD.0000000000031030 [RCT, double-blind, n=83]
- Kamal S, Abdelhakam S, Ghoraba D et al. The frequency, clinical course, and health related quality of life in adults with Gilbert's syndrome: a longitudinal study. BMC Gastroenterol. 2019;19(1):22. DOI: 10.1186/s12876-019-0931-2 [Cohort study, longitudinal, n=101]
- Roager HM, Hansen LBS, Bahl MI et al. Colonic transit time is related to bacterial metabolism and mucosal turnover in the gut. Nat Microbiol. 2016;1(9):16093. DOI: 10.1038/nmicrobiol.2016.93 [Human observational study, metabolomics]
- Goldin BR, Adlercreutz H, Gorbach SL et al. Estrogen excretion patterns and plasma levels in vegetarian and omnivorous women. N Engl J Med. 1982;307(25):1542-1547. DOI: 10.1056/NEJM198212163072502 [Human comparative study, n=20, four measurement points]
- Gorbach SL, Goldin BR. Diet and the excretion and enterohepatic cycling of estrogens. Prev Med. 1987;16(4):525-531. DOI: 10.1016/0091-7435(87)90067-3 [Human comparative study, several cohorts]
- Cohn WJ, Boylan JJ, Blanke RV, Fariss MW, Howell JR, Guzelian PS. Treatment of chlordecone (Kepone) toxicity with cholestyramine. Results of a controlled clinical trial. N Engl J Med. 1978;298(5):243-248. DOI: 10.1056/NEJM197802022980504 [Controlled clinical trial, n=22]
- Boylan JJ, Cohn WJ, Egle JL, Blanke RV, Guzelian PS. Excretion of chlordecone by the gastrointestinal tract: evidence for a nonbiliary mechanism. Clin Pharmacol Ther. 1979;25(5 Pt 1):579-585. DOI: 10.1002/cpt1979255part1579 [Follow-up by the same research group, human and animal model]
- American Academy of Clinical Toxicology, European Association of Poisons Centres and Clinical Toxicologists. Position statement and practice guidelines on the use of multi-dose activated charcoal in the treatment of acute poisoning. J Toxicol Clin Toxicol. 1999;37(6):731-751. DOI: 10.1081/clt-100102451 [Position statement and guideline, professional societies]
- Jacobson TA, Armani A, McKenney JM, Guyton JR. Safety considerations with gastrointestinally active lipid-lowering drugs. Am J Cardiol. 2006;99(6A):47C-55C. DOI: 10.1016/j.amjcard.2006.11.022 [Review, drug safety]
- Gao S, Sun R, Singh R et al. The role of gut microbial β-glucuronidase in drug disposition and development. Drug Discov Today. 2022;27(10):103316. DOI: 10.1016/j.drudis.2022.07.001 [Review. Background reading, not cited individually in the text]
- Wang P, Jia Y, Wu R, Chen Z, Yan R. Human gut bacterial β-glucuronidase inhibition: An emerging approach to manage medication therapy. Biochem Pharmacol. 2021;190:114566. DOI: 10.1016/j.bcp.2021.114566 [Review. Background reading, not cited individually in the text]
- Ioannides C, Konsue N. A principal mechanism for the cancer chemopreventive activity of phenethyl isothiocyanate is modulation of carcinogen metabolism. Drug Metab Rev. 2015;47(3):356-373. DOI: 10.3109/03602532.2015.1058819 [Mechanism review, phase I and phase II]
Transparency about the evidence. The mechanism described in this article, that is conjugation in the liver, biliary release, bacterial removal of the tag in the gut and possible re-uptake, is very well documented at the biochemical level. The most convincing evidence for its clinical relevance in humans comes from drug therapy, for example on mycophenolate and on bilirubin, and from a controlled study in people with chlordecone poisoning. A large part of the mechanistic data, by contrast, comes from cell experiments and animal models. For the question of whether the body burden of everyday environmental exposures in healthy people can be measurably lowered by fibre, binders or microbiome-related measures, no robust randomised studies exist to my knowledge. The relationships between diet, stool weight and estrogen excretion come from comparative studies, not from randomised interventions. The statements about beta-glucuronidase are also not to be read in one direction: a low activity was found in one study in a damaged microbiome as well.
No substitute for medical advice. This article is for information and does not replace an individual medical examination, diagnosis or treatment. Persistently changed bowel habits, blood in the stool, unexplained weight loss, newly appeared yellowing of skin or eyes as well as severe or persistent abdominal pain belong in a medical assessment. Binders such as activated charcoal or bile acid sequestrants can reduce the uptake of medicines and nutrients and are contraindicated in certain situations. Anyone who takes medicines regularly, in particular thyroid hormones, anticoagulants, immunosuppressants or hormonal contraception, anyone who is pregnant or breastfeeding, anyone who has a liver disease, gallstone disease or another biliary tract disease, or a chronic inflammatory bowel disease, should discuss the use of such agents with a doctor beforehand. A change to a medically prescribed diet, for example a low-fat diet, also belongs in that conversation beforehand. An existing therapy should never be reduced or stopped on your own.