Are lectins and antinutrients bad for you?

The objection

“Plants make lectins, oxalates and phytates to stop animals eating them, and those defences don’t switch off because the plant is on your plate. As few as four or five raw kidney beans can make you violently ill. In animal studies, lectins from raw beans damaged the lining of the gut and stunted growth, phytate binds iron and zinc so that you absorb less of them, and oxalate is what most kidney stones are made of. Why build a diet on foods that have to be detoxified before they’re safe to eat?”

The answer

The chemistry in that objection is right, and what it leaves out is the kitchen. Lectins are proteins that bind to sugars on the surface of cells, found in almost all living things, animals included, and plants make them partly as a defence against insects, moulds and disease (Petroski and Minich, 2020) [1]. Raw pulses and wholegrains are far more concentrated sources than fruit and vegetables, which supply insignificant amounts even eaten raw [1], something worth knowing if you’re weighing up a raw vegan diet.

Soaking on its own does little to lectins, but cooking inactivates the overwhelming majority of them. A University of Manitoba team measured both steps in 22 kinds of Canadian peas, lentils, fava (broad) beans, chickpeas and beans, with soya beans for comparison (Shi, Arntfield and Nickerson, 2018) [2][3]. Raw soya had by far the most lectin activity, 692.8 units per milligram, followed by the beans at about 88 and the other pulses at between 2.7 and 11 [2]. Four hours of soaking at room temperature cut lectin activity by no more than about 5% [2][3]. Soaking followed by an hour in water at 95 °C, just below boiling point, cut it by between 93.8% and 99.8% in every pulse tested, according to the full tables in the thesis the study came from, and in dark red kidney beans it fell from 88.5 units to 0.17 (Shi, 2015) [1][3]. Soaking did far more for soluble oxalate than for lectins, cutting it by between 27% and 56% [2]. After soaking and cooking it was between about a third and three-quarters lower than in the raw seeds, though the study found no significant difference between the soaked and the cooked seeds [3].

That runs against advice you’ll often read. Harvard’s Nutrition Source says soaking in water for several hours can inactivate most lectins [4], and longer soaks may do more than four hours did here: the thesis cites a mung bean study in which a 12-hour soak cut lectin activity by about half [3]. But the US Food and Drug Administration says kidney bean poisoning is usually caused by eating raw, soaked beans (FDA, 2012) [5]. Its figures for kidney beans run from 20,000 to 70,000 haemagglutinating units raw to 200 to 400 when fully cooked [5].

How do you cook dried kidney beans safely?

Soak dried kidney beans for at least five hours, pour the water away and boil them hard in fresh water for at least 30 minutes, which meets both the Australian and the US food safety advice [5][6]. Australia’s Food Safety Information Council (2023) gives the soaking advice and says to boil them briskly for at least ten minutes [6]. The FDA (2012) reports that ten minutes at 100 °C completely destroyed the toxin in tests, but tells consumers to boil the beans for at least 30 minutes so that they get hot enough for long enough [5]. Both say not to cook them in a slow cooker [5][6]. In studies of casseroles made in slow cookers the food often reached only 75 °C or less, too little to destroy the toxin [5], and the council warns that undercooked beans can be more toxic than raw ones [6]. Eating as few as four or five raw beans can bring on severe vomiting and diarrhoea, usually within one to three hours, and while most people recover within three or four hours, some have needed hospital treatment [5][6]. In the UK, 50 suspected incidents of red kidney bean poisoning were reported between 1976 and 1989, nine of them confirmed by finding the lectin in the beans, and in many the beans had been eaten raw or after inadequate heating (Rodhouse et al., 1990) [7]. Tinned beans are already cooked and can be eaten without further cooking [4][6].

Do lectins damage your gut?

In the general population there is no good evidence that properly cooked beans and grains damage the gut [1]. The studies showing harm used isolated lectins or raw bean flour, mostly in animals, where high doses altered the gut lining, made it more permeable and slowed growth, and the few human trials with whole cooked beans have not shown the same effects (Petroski and Minich, 2020) [1]. That review found no strong evidence from human trials that lectin-rich foods, whole and cooked, consistently cause inflammation, a leaky gut or problems absorbing nutrients in the general population [1], and Harvard’s Nutrition Source points out that research in humans on the lectins people actually eat is very limited [4]. People with a sensitive gut, such as those with irritable bowel syndrome, may be more likely to get symptoms, and Harvard suggests eating less of, or less often, whatever causes the trouble [4]. Coeliac disease is a separate condition: it is an immune reaction to gluten, a protein in wheat, barley and rye, and it is managed with a gluten-free diet (NHS, 2023) [18]. Harvard also notes that the theory has fed a profitable anti-lectin movement of bestselling books and enzyme supplements [4].

Does phytate stop you absorbing iron and zinc?

Phytate does reduce how much iron and zinc you absorb from a meal, and it survives cooking far better than lectins do. It binds iron, zinc and calcium in the gut, though only when it is eaten at the same meal as those minerals [8]. In the Canadian study, four hours of soaking made no detectable difference to it, and cooking left it almost unchanged in kidney, pinto, navy and black beans and in soya, though it fell in peas, lentils, fava beans and chickpeas [2][3]. In dark red kidney beans, the lectin activity fell by 99.8% while the phytate was 2.3% lower, a difference the study could not distinguish from no change [3]. Sprouting and fermenting can do more: in studies summarised in the 2020 review, germinating chickpeas and pigeon peas cut their phytate by more than 60%, and the natural leavening of bread also reduces it significantly [1].

The British Dietetic Association notes that iron and zinc are absorbed less well from plant foods than from animal foods (BDA, 2021) [9]. For iron it advises plenty of fruit and vegetables rich in vitamin C, and for zinc it suggests fermented soya such as tempeh and miso, soaking dried beans and rinsing them before cooking, and sprouting grains and seeds [9]. Of those, sprouting and fermenting are shown to cut phytate in the studies cited here [1], while soaking is less clear-cut. The four-hour soak in the Canadian study made no detectable difference to it [2], and in a 2005 study summarised in the 2020 review, soaking millet, maize, rice and soya beans cut their phytate by between 17% and 28% but washed iron and zinc out into the water as well, so it did not lower the ratio of phytate to iron and changed the ratio to zinc only slightly [1]. Harvard’s Nutrition Source names vegans, along with people already at risk of running short of these minerals, among those who may need to think about phytate [8]. It also reports laboratory and animal studies in which phytate protected against DNA damage and cancer cell growth, but stresses that the research is early [8], and results like these prove no more about benefit in people than the animal lectin studies prove about harm.

The clearest sign that lower absorption might matter comes from children. A 2024 meta-analysis of 18 studies, 17 of them cross-sectional, found that vegan children and adolescents ate more iron than omnivorous ones yet had lower ferritin, the body’s iron store, though every pooled result in it was rated low or very low certainty (Koller et al., 2024) [16]. A 2026 meta-analysis of 59 studies covering 48,626 children and adolescents, 1,289 of them vegan, found lower haemoglobin and ferritin in the vegans, while noting that group averages for most nutrients and blood markers stayed within the reference ranges for children (Lotti et al., 2026) [17]. Neither can say how much of the gap is down to phytate, but the 2026 authors call for careful dietary planning and supplementation in lacto-ovo-vegetarian and vegan children [17], so plan a vegan child’s diet with a GP or dietitian, who can check iron levels and advise on supplements.

Should you avoid oxalates?

Most people have no reason to avoid oxalates in ordinary meals, but anyone who has had calcium oxalate kidney stones should follow their doctor’s advice, and anyone with kidney disease or a past gastric bypass should talk to their doctor before trying a juice cleanse or drinking large green smoothies every day [14]. Most kidney stones are made of calcium oxalate, but the oxalate in food turned out to be a modest factor in three large US cohorts of 240,681 health professionals and nurses: those who ate the most had a 22% higher risk of stones among men and 21% among older women than those who ate the least, and no higher risk among younger women (Taylor and Curhan, 2007) [10]. Spinach supplied more than 40% of the oxalate they ate, and the authors concluded that their data did not implicate dietary oxalate as a major risk factor [10]. In the same cohorts, people whose diets scored highest for fruit, vegetables, nuts, pulses, wholegrains and low-fat dairy, and lowest for salt, sugary drinks and red and processed meat, had a 40 to 45% lower risk of stones than those who scored lowest, despite eating more oxalate (Taylor, Fung and Curhan, 2009) [11]. Both are observational studies, and the figures are relative risks.

Very large doses are another matter. Case reports describe kidney failure in a 65-year-old woman whose kidneys had worked normally before she drank a green smoothie juice cleanse, and who had had a gastric bypass years earlier and a long course of antibiotics more recently (Makkapati, D’Agati and Balsam, 2018) [14], and in a 54-year-old man with type 2 diabetes and chronic kidney disease who regularly drank spinach and kale smoothies and was still on dialysis when the report was written (Samra and Gupta, 2024) [15]. These are single cases, and Makkapati and colleagues name chronic kidney disease, gastric bypass and antibiotic use as risk factors that patients and doctors should know about [14].

Cooking helps here too. Boiling vegetables for 12 minutes cut their soluble oxalate by between 30% and 87% in one study, with the largest losses in spinach and Swiss chard, while steaming did less [1][8]. Eating calcium-rich food at the same meal, such as calcium-fortified plant milk with beans, spinach or nuts, binds the oxalate before it can be absorbed [8]. Harvard adds that some people prone to stones may absorb more oxalate than usual, and for them eating less of the foods richest in oxalate may be particularly helpful [8]. The NHS advice for preventing stones is to drink 2.5 to 3 litres of water a day as an adult, get enough calcium and have less than 6 g of salt a day, and it lists a high-protein, low-fibre diet among the things that make stones more likely to come back (NHS, 2026) [12]. It also says not to change your diet without speaking to your doctor [12], so if you’ve had kidney stones, ask your GP or a dietitian before cutting down on high-oxalate foods, and don’t change or stop any medicine without speaking to your GP.

Do people who eat the most beans live longer?

In cohort studies, people who eat the most legumes die at slightly lower rates overall, and none of the pooled results points to them dying sooner. In a 2023 meta-analysis of 32 cohorts covering more than a million people, the 27 cohorts that reported deaths from all causes found about 6% lower mortality among those who ate the most legumes than among those who ate the least, which is hard to square with antinutrients doing serious harm at the amounts people actually eat (Zargarzadeh et al., 2023) [13]:

Deaths from (cohorts)	Hazard ratio (95% CI)	Certainty
All causes (27)	0.94 (0.91 to 0.98)	Moderate
Stroke (5)	0.91 (0.84 to 0.99)	Moderate
Cardiovascular disease (11)	0.99 (0.91 to 1.09)	Low
Coronary heart disease (5)	0.93 (0.78 to 1.09)	Low
Cancer (5)	0.85 (0.72 to 1.01)	Low

A hazard ratio below 1 means fewer deaths among the highest legume eaters, and a range that crosses 1 means the analysis could not rule out no difference. These are relative figures from observational studies, the paper gives no absolute differences, and its authors marked the certainty down mainly for possible residual confounding and inconsistency between studies [13]. The stroke result was not robust and depended on two studies, and the lower all-cause death rate showed up in cohorts from Asia and Australia and in two international ones, while those in the US (hazard ratio 1.00) and Europe (1.01) found no difference either way [13]. None of the five pooled estimates was above 1, though the ranges for cardiovascular and coronary deaths reach 1.09 and the one for cancer 1.01, so for those three a small increase can’t be ruled out any more than a benefit can [13].

If cutting out foods, or worrying about what’s in them, starts to take over, eating-disorder charities can help: Beat in the UK, Bodywhys in Ireland, ANAD in the US, NEDIC in Canada, the Butterfly Foundation in Australia, and EDANZ, which supports families and carers, in New Zealand.