Does glyphosate exposure cause fatty liver disease in humans, and what do the human studies show about urinary glyphosate and liver outcomes?

The short version: no human study has proven glyphosate causes fatty liver disease, but the human data that exists points the same direction as the animal data, and the question has never been properly studied.

The human evidence is thin and indirect. One large US survey of about 5,200 adults found that higher urinary glyphosate tracked with a higher metabolic syndrome score, which includes the blood-sugar and cholesterol problems that travel with fatty liver [3]. The pattern was not a straight line: the strongest effect appeared at the third quartile of exposure, not the highest, and it was more pronounced in older adults and in Mexican-American, Other Hispanic, and Black participants [3]. That is a correlation in a snapshot study, not proof of cause. It cannot tell you whether glyphosate arrived before the liver disease or after it.

The rest of the human record is weaker still. A 2004 study of 48 farm families found farmers had a geometric mean urine level of about 3 parts per billion on application day, with children mostly undetectable, and concluded doses were far below the EPA reference value [2]. A 2021 study of 41 Portuguese children found glyphosate in 95% of urine samples, mean about 1.8 micrograms per liter, with higher levels in kids aged 7 to 9 and those living near farms [4]. Neither study looked at liver outcomes at all. One 2018 case report described a 42-year-old man with transient liver enzyme elevations after acute glyphosate poisoning, with recovery over about six months [6]. That is a single case, not epidemiology.

The animal data fills in the mechanism the human studies cannot see. Rats exposed to ultra-low-dose Roundup developed non-alcoholic fatty liver disease, confirmed by multiomics [1]. Pregnant rats given glyphosate showed sharply increased liver lipid peroxidation, especially in fetal livers, where the marker roughly quadrupled [7]. Another rat study found low-dose glyphosate and Roundup disrupted the urine metabolome and gut microbiome, with elevated homocysteine in male pups [5]. These are consistent with a liver under oxidative stress, which is the same pathway that drives fatty liver disease in humans.

What is missing is the study that would settle this: a prospective human cohort that measures glyphosate exposure before liver disease develops, with imaging or biopsy to confirm fatty liver, and that controls for diet, alcohol, and obesity. That study has not been done, or at least not surfaced in this retrieval. The cross-sectional design cannot establish causation, and no retrieved study directly measured fatty liver against glyphosate levels in humans. The burden of proof sits with those who say glyphosate is harmless to the liver at real-world exposure, and that burden has not been met.

My call: the human evidence is suggestive but not conclusive, the animal mechanism is coherent, and the definitive human study has never been run. Confidence: low.