Are Emulsifiers Bad for You? What the Measurements Show

Before anyone tells you a chemical strips your gut wall, somebody has to measure the wall. The figure in wide circulation is roughly 830 µm of mucus gel in the colon, with about 116 µm of it firmly adherent. Those are rat numbers, taken in anesthetized animals. Type “are emulsifiers bad for you” into a search bar and you will meet that figure within two clicks, attached to a human body it was never measured in.

The human measurement is a fraction of it. On Carnoy-fixed sigmoid biopsies in a colonoscopy cohort, the mean inner mucus thickness was 19.4 µm, standard deviation 9.9, with a range from 6.2 to 45.6. The loose outer layer was absent from every biopsy, because bowel preparation strips it. My video Your Microbes Read the Label, Not the Macros covers the chemistry. Below are the measurement problem and the repair question it leaves.

The wall everybody quotes is a rat, and the jar had a vote

In paired samples from the same animals, the choice of fixative alone changed the measured inner mucus thickness, with a p below 0.0001, and changed the count of bacteria sitting next to the epithelium, at p below 0.0022. What did not change was everything that is not a gel: mucosal height ran p = 0.42 and goblet cell number p = 0.66. The tissue stayed the same. Only the jar changed, and only the soft thing in the tissue noticed.

So every human mucus number came out of a container, not a living colon. It cuts the other way too: the layer renews fast. Human colonic biopsies in a perfusion chamber grew mucus at about 240 microns an hour, and one anchor review describes the inner layer as renewed hourly by surface goblet cells. Whatever happened to that wall yesterday, your body is already building the replacement.

Are emulsifiers bad for you? The mechanism runs through your bacteria

Here is the finding that should have ended the dish-soap explanation. In germ-free mice, emulsifiers did not thin the mucus and did not move bacteria-sized beads any closer to the wall. In animals carrying a restricted defined flora, neither compound did anything at all. In a human microbial community cultured ex vivo, cellulose gum raised the community’s inflammatory potential without significantly altering which species were present.

Thickness is not the measurement that matters

The most useful correction in this whole literature is that thickness and barrier function are different readings. Human sigmoid biopsies from healthy controls showed clear separation between the epithelium and both bacteria and beads. In active ulcerative colitis the mucus was highly penetrable. In remission, most patients had an impenetrable layer like controls. And six mouse colitis models all showed bacteria touching the epithelium, including interleukin-10-deficient animals whose layer was thicker than wild type and leaked anyway.

A thicker wall that leaks is still a wall that leaks. Measure the leak.

In a proteomics study, abnormal penetrability appeared in only 30% of patients with active colitis, so not everyone with the disease carries the defect. The sharpest check on that story is an aging experiment: littermate mice at nineteen months had a colonic mucus layer about 6-fold thinner than mice at ten weeks, absent or one to five microns where young animals showed a gap of about 50 microns, with bacteria chronically against the epithelium. There was no histological evidence of colitis. A destroyed barrier did not produce a disease. Barrier failure is a mechanism, not a verdict, and that distinction is what keeps this from becoming another panic marker.

What a stool sample cannot see

Mice fed cellulose gum or polysorbate 80 showed stark changes in the microbes living in the mucus, and those changes were markedly distinct from the changes in feces. Transfer only the mucus microbiota into germ-free animals and the recipients developed encroachment, low-grade inflammation and features of metabolic syndrome.

That is mouse work with no human version. But human trials sequence stool, and if the action sits in the mucus, a stool census is looking in the wrong room.

Human encroachment has been measured once, and what predicted it is the interesting part. In 42 people undergoing colonoscopy, the distance from the nearest bacteria to the nearest epithelial cell was reduced almost 3-fold in patients with type 2 diabetes. The r² ran 0.16 for body mass index, 0.46 for fasting glucose and 0.51 for hemoglobin A1c. Obesity by itself did nothing, and removing the subjects with diabetes eliminated the difference completely. The wall tracked the host’s blood sugar, not the waistline, which is the whole argument for treating metaflammation as the terrain rather than chasing one ingredient.

The trial that removed the host, and the one that kept it

One randomized controlled feeding trial has given a purified emulsifier to healthy people. Sixteen adults ate an otherwise emulsifier-free diet for eleven days, and seven of them had 15 g a day of cellulose gum added. The primary measure, bacterial-epithelial distance, did not change; encroachment appeared in 2 of 7 treated and 0 of 9 controls. Fecal lipocalin-2 did not move, at p = 0.258, and neither did fecal lipopolysaccharide, flagellin or serum cytokines. Postprandial abdominal pain did rise.

Its power calculation explains the null: it was built to detect the 19.13-micron gap measured between people with and without diabetes. It was powered to find a diabetic-sized effect in volunteers who did not have diabetes.

The trial that kept sick patients in found something. All 154 adults with active Crohn’s disease started on restricted intake, and one arm had emulsifiers returned under blinding. Beyond the symptom response, the adjusted risk ratio for at least halving fecal calprotectin was 2.9 (95% CI 1.1-8.0). Calprotectin is not a feeling; it is a number your gastroenterologist already orders. It is also a congress abstract, with 113 of 154 finishing, and it should be weighed as one.

What actually repairs the terrain

Two mechanisms arrive in the same mouthful, and only one of them gets argued about. In gnotobiotic mice, fiber deprivation, even on alternating days, drove the microbiota to degrade host mucus, thinning the layer five- to six-fold, and purified prebiotic fiber did not prevent it. The fuel those bacteria are supposed to be eating is the other half of the problem.

Then the feeding trials, which are the closest thing to a repair signal anyone has. Fifty adults ate two eight-week diets in random order, both built to meet national dietary guidance, one ultra-processed and one minimally processed. Both produced weight loss; minimally processed gave 2.06% against 1.05%, a difference of 1.01 percentage points at p = 0.024, with fat mass falling 0.98 kg further and craving control improving more. Be fair about what did not move: waist circumference did not differ, LDL cholesterol was actually lower on the ultra-processed diet, and C-reactive protein did not change from baseline on either.

In a second trial, 36 older adults who cut ultra-processed intake from about 50% of energy to about 13% improved insulin, apolipoprotein B and C-reactive protein, then drifted back toward baseline as intake climbed to 44% at one year. The diet was not wrong. It stopped.

So the prescription is not a supplement and not a purge. It is fewer items with ingredient lists and more with none, sustained, because the terrain repairs on a supply that keeps arriving. The tier is mechanism, inflamed-host animal work and one abstract, plus feeding trials that improved almost everything whenever the shelf came out, whether or not the emulsifier was the reason. A body in that state answers interventions differently, which is why we ask about the label before we plan a treatment. Every study behind this, with its numbers and its limits, is in the companion Deep Dive. Next comes the immune system that meets whatever crosses that layer.

Frequently asked questions

Can a test tell me whether my gut barrier is leaking?

Not reliably, and the honest answer matters more than a number. Human mucus has only been measured in biopsies, where the fixative alone changes the reading, and the animal work says the action sits in the mucus compartment that stool sequencing cannot reach. Encroachment has been measured in research colonoscopies, not in commercial panels. Markers have to earn their place in a workup.

Does blood sugar change how my gut wall behaves?

It tracked with it more closely than weight did. In 42 people undergoing colonoscopy, bacterial distance from the epithelium was reduced almost 3-fold in type 2 diabetes, with r² of 0.46 for fasting glucose and 0.51 for hemoglobin A1c, while obesity alone did nothing and removing the diabetic subjects erased the difference. Glucose control shows up in tissues nobody associates with it.

If I cut processed food, what improves and what does not?

In an eight-week crossover, the minimally processed diet gave 2.06% weight loss against 1.05%, more fat loss and better craving control, while waist circumference did not differ and C-reactive protein did not change from baseline. Duration is doing most of the work. Metabolism sets the ceiling on repair.

Is this the same idea as leaky gut?

It overlaps, and the difference is discipline. Barrier penetrability is measurable, appeared in 30% of patients with active colitis, and aged mice lost almost their whole mucus layer with no colitis at all, so a thin wall is not automatically a disease. We treat it as one input to a systemic inflammatory picture instead of a diagnosis on its own. The gut and the joint share more than most workups admit.

Sources

1. Atuma, C., Strugala, V., Allen, A., & Holm, L. (2001). The adherent gastrointestinal mucus gel layer: thickness and physical state in vivo. American Journal of Physiology. Gastrointestinal and Liver Physiology, 280(5), G922-G929. https://doi.org/10.1152/ajpgi.2001.280.5.G922
2. Jawhara, M., Sorensen, S. B., Heitmann, B. L., Halldorsson, T. I., Pedersen, A. K., & Andersen, V. (2020). The relation between red meat and whole-grain intake and the colonic mucosal barrier: A cross-sectional study. Nutrients, 12(6), 1765. https://doi.org/10.3390/nu12061765
3. Blick, A. K., Giaretta, P. R., Sprayberry, S., Bush-Vadala, C., Paulk, C. B., Boeckman, J., Callaway, T. R., Gill, J. J., & Rech, R. R. (2019). Comparison of 2 fixatives in the porcine colon for in situ microbiota studies. Journal of Animal Science, 97(12), 4803-4809. https://doi.org/10.1093/jas/skz325
4. Chassaing, B., Raja, S. M., Lewis, J. D., Srinivasan, S., & Gewirtz, A. T. (2017a). Colonic microbiota encroachment correlates with dysglycemia in humans. Cellular and Molecular Gastroenterology and Hepatology, 4(2), 205-221. https://doi.org/10.1016/j.jcmgh.2017.04.001
5. Chassaing, B., Compher, C., Bonhomme, B., Liu, Q., Tian, Y., Walters, W., Nessel, L., Delaroque, C., Hao, F., Gershuni, V., Chau, L., Ni, J., Bewtra, M., Albenberg, L., Bretin, A., McKeever, L., Ley, R. E., Patterson, A. D., Wu, G. D., … Lewis, J. D. (2022). Randomized controlled-feeding study of dietary emulsifier carboxymethylcellulose reveals detrimental impacts on the gut microbiota and metabolome. Gastroenterology, 162(3), 743-756. https://doi.org/10.1053/j.gastro.2021.11.006
6. Johansson, M. E. V., Gustafsson, J. K., Holmen-Larsson, J., Jabbar, K. S., Xia, L., Xu, H., Ghishan, F. K., Carvalho, F. A., Gewirtz, A. T., Sjovall, H., & Hansson, G. C. (2014). Bacteria penetrate the normally impenetrable inner colon mucus layer in both murine colitis models and patients with ulcerative colitis. Gut, 63(2), 281-291. https://doi.org/10.1136/gutjnl-2012-303207
7. van der Post, S., Jabbar, K. S., Birchenough, G., Arike, L., Akhtar, N., Sjovall, H., Johansson, M. E. V., & Hansson, G. C. (2019). Structural weakening of the colonic mucus barrier is an early event in ulcerative colitis pathogenesis. Gut, 68(12), 2142-2151. https://doi.org/10.1136/gutjnl-2018-317571
8. Sovran, B., Hugenholtz, F., Elderman, M., Van Beek, A. A., Graversen, K., Huijskes, M., Boekschoten, M. V., Savelkoul, H. F. J., De Vos, P., Dekker, J., & Wells, J. M. (2019). Age-associated impairment of the mucus barrier function is associated with profound changes in microbiota and immunity. Scientific Reports, 9(1), 1437. https://doi.org/10.1038/s41598-018-35228-3
9. Kordahi, M. C., Daniel, N., Gewirtz, A. T., & Chassaing, B. (2025). Mucus-penetrating microbiota drive chronic low-grade intestinal inflammation and metabolic dysregulation. Gut Microbes, 17(1), 2455790. https://doi.org/10.1080/19490976.2025.2455790
10. Dicken, S. J., Jassil, F. C., Brown, A., Kalis, M., Stanley, C., Ranson, C., Ruwona, T., Qamar, S., Buck, C., Mallik, R., Hamid, N., Bird, J. M., Brown, A., Norton, B., Gandini Wheeler-Kingshott, C. A. M., Hamer, M., van Tulleken, C., Hall, K. D., Fisher, A., Makaronidis, J., & Batterham, R. L. (2025). Ultraprocessed or minimally processed diets following healthy dietary guidelines on weight and cardiometabolic health: A randomized, crossover trial. Nature Medicine, 31(10), 3297-3308. https://doi.org/10.1038/s41591-025-03842-0

Medically reviewed by Dr. Gurpreet Singh Padda, MD, MBA, MHP, board certified in anesthesiology, pain medicine, interventional pain management, addiction medicine and obesity medicine. Adapted from Chapter 15 of The Angry Gut by Dr. Padda and Ami Michelle Grimes. Last reviewed September 2026.