Vegetables: what the pot changes, and what is already on them

Cooking is the part of a vegetable you control, and it moves the numbers further than most people expect. Pressure cooking pulled spinach down to 74.85 milligrams of potassium per 100 grams and courgette to 93.62, starting from a list where every food held more than 201. For someone with kidney disease told to keep potassium down, that is the difference between a vegetable being off the list and being on it. It is also a question for the renal dietitian rather than a decision to make from a web page. In the other direction, among the ordinary ways to cook a cabbage-family vegetable, steaming preserved the most glucosinolates, the compounds broccoli and kale are studied for, and boiling and blanching preserved less. The composition table drifts as well. Tracking Australian vegetables back to 1900, iron fell 30 to 50 percent in sweet corn, red-skinned potatoes, cauliflower and green beans, and rose 150 to 300 percent in Hass avocados, mushrooms and silverbeet.

The numbers, per 100 g

How much is actually in it

• Pesticide residues turned up four times as often on conventional crops, and cadmium was higher in them too. — 343 peer-reviewed publications (n = 343 studies) Evidence C source
  Phenolic acids were an estimated 19% higher in organic crops (95% CI 5 to 33), flavanones 69% (13 to 125), stilbenes 28% (12 to 44), flavones 26% (3 to 48), flavonols 50% (28 to 72) and anthocyanins 51% (17 to 86); pesticide residues occurred four times more often in conventional crops, which also carried significantly more cadmium; the authors link both to specific agronomic practices such as not using mineral nitrogen and phosphorus fertilisers

What cooking and storage change

• Of the ordinary ways to cook a cabbage-family vegetable, steaming kept the most of the compound people cook it for. — over 100 standardised articles on Brassica species and varieties (n = 100+ articles) Evidence C source
  The highest glucosinolate content was in Brassica nigra at 201.95 ± 53.36 µmol/g followed by the Alboglabra group of B. oleracea at 180.9 ± 70.3; across boiling, blanching and steaming, steaming was judged most favourable for preserving glucosinolates and their derivatives, and the authors recommend a mild-processing approach

Safety, limits and interactions

• A country-wide audit of pesticide residues found sixteen above the legal limit and still calculated the dietary risk as not severe. — 26 fruit and vegetable sample types across Brazilian states, for children and adults (n = 33 articles) Evidence C source
  111 residues were identified, mainly organophosphates and organochlorines, of which 16 exceeded the maximum residue limit; no severe acute or chronic dietary risks were found, with the highest values in São Paulo and Santa Catarina from organophosphates in tomatoes and sweet peppers; 26 of the residues are classed as carcinogenic by the US EPA and the cumulative hazard index stayed below 1 in every state
• Where produce is irrigated with untreated water, more than half of the leafy samples carried parasites. — 29 studies covering 6659 vegetable and fruit samples (n = 6659 samples) Evidence C source
  Overall contamination was 48.76% (95% CI 40.78 to 56.74), with bacterial contamination 52.85% and parasitic 46.40%; lettuce at 61.0%, salad at 57.4%, spinach at 54.2% and cabbage at 52.9% were the most parasite-contaminated; the commonest parasites were Ascaris, Entamoeba, Strongylida and Giardia

What people believe that the data does not support

• Iron in Australian vegetables fell by a third to a half in four of them and rose by up to triple in three others. — 34 articles and six versions of national food composition databases (n = 34 articles) Evidence C source
  Most vegetables showed similar iron across two or more time points; sweet corn, red-skinned potatoes, cauliflower and green beans fell 30 to 50%; Hass avocados, mushrooms and silverbeet rose 150 to 300%; legumes showed more pronounced reductions with higher and more variable values before 2000
• US law permits the claim that vegetables lower cancer risk and states in the same rule that nobody agrees on why. — US food labels Evidence E source
  The rule states that many studies show diets high in plant foods are associated with reduced risk of some cancers, and that there is not scientific agreement as to whether the protective effect comes from the nutrient components including fibre, beta-carotene and vitamin C, from the displacement of fat from such diets, or from other substances in these foods which are not nutrients

Who this works differently for

• Pressure cooking pulled spinach below the potassium threshold that puts a vegetable off the list for kidney patients. — vegetables and legumes from the Brazilian TACO table, cooked by different techniques, with nobody's blood potassium measured. A separate phase asked 68 patients at a cardiology institute to rate how acceptable the food was (n = 68) Evidence B source
  The lowest potassium values were spinach at 74.85 mg per 100 g and courgette at 93.62 mg, both pressure-cooked; among the low values only squash at 237.83 mg stayed above 200 mg; beans fell below the threshold under every technique tested

What is still unknown

• The organic comparison pooled crops, countries and climates together because there was no other way, and its heterogeneity shows it. — the same 343 publications (n = 343 studies) Evidence B source
  Heterogeneity exceeded I² of 75% for most composition parameters where significant differences were found; the authors state that pooling diverse information was necessary because too few studies existed to analyse specific countries, regions, crops and species separately; for antioxidant and phenolic parameters they rate reliability high or moderate for 13 of 19, and moderate for cadmium

The bottom line

The organic question has a measured answer and a wide error bar. Pesticide residues turned up four times as often on conventional crops, which also carried significantly more cadmium, and organic crops ran higher on several groups of plant compounds, from 19 percent for phenolic acids to 51 percent for anthocyanins. The same authors say plainly why you should hold that loosely. They had to pool crops, countries, soils and climates together because too few studies existed to look at any one of them, and heterogeneity ran above 75 percent for most of the parameters where a difference showed up. Whether the residue difference matters to you is a separate question from whether it exists. A country-wide Brazilian audit found 111 residues, 16 of them over the legal limit, and still calculated no severe acute or chronic dietary risk, with the cumulative hazard index below 1 in every state. Washing matters most where the water does. In Ethiopia, where produce is often irrigated with untreated water, 48.76 percent of samples carried bacterial or parasitic contamination, and lettuce came in at 61 percent. That number belongs to those growing conditions and does not transfer to a supermarket somewhere else. One last thing, about the claim on the packet. American law permits a food label to say that diets high in fruits and vegetables reduce cancer risk. The same rule states there is no scientific agreement on where that effect comes from, listing fibre, beta-carotene and vitamin C, the displacement of fat, and other substances in these foods that are not nutrients. The regulator is telling you it does not know the mechanism, on the same page where it lets the claim be printed. Back to the potassium, because it is the one number here where getting it wrong costs something the same week. That trial measured potassium in food, cooked different ways. It measured nobody's blood, and no patient was followed. If you have kidney disease and have been told to keep potassium down, the technique is worth taking to the renal dietitian who has your numbers in front of them. It is not worth acting on from a page.

Sources

1. veg-r5-04 · systematic review and meta-analysis
   Additionally, the frequency of occurrence of pesticide residues was found to be four times higher in conventional crops, which also contained significantly higher concentrations of the toxic metal Cd.
   Br J Nutr, 2014 · checked 2026-09-22
2. veg-r5-02 · systematic review
   The authors also conclude that food processing can influence significantly the final content of GLS, considering the most popular methods: boiling, blanching, steaming, the latter can be considered as the most favorable to preserve highest level of GLS and their deriviatives.
   Crit Rev Food Sci Nutr, 2021 · checked 2026-09-22
3. veg-r5-07 · systematic review with health risk assessment
   We did not identify severe acute and chronic dietary risks, but the highest risk values were observed in São Paulo and Santa Catarina, associated with the consumption of tomatoes and sweet peppers due to the high concentrations of organophosphates.
   Chemosphere, 2023 · checked 2026-09-22
4. veg-r5-08 · systematic review and meta-analysis
   Lettuce (61.0%), salad (57.4%), spinach (54.2%), and cabbage (52.9%), were the vegetables most contaminated by parasites
   BMC Infect Dis, 2025 · checked 2026-09-22
5. veg-r5-03 · scoping review
   The majority of vegetables had similar iron content between two or more timepoints but decreases of 30-50% were noted for sweet corn, red-skinned potatoes, cauliflower and green beans while increases of 150-300% were seen for Hass avocadoes, mushrooms and silverbeet.
   Foods, 2021 · checked 2026-09-22
6. veg-r5-06 · codified federal regulation
   Currently, there is not scientific agreement as to whether the observed protective effects of fruits and vegetables against cancer are due to a combination of the nutrient components of diets rich in fruits and vegetables, including but not necessarily limited to dietary fiber, vitamin A (as beta-carotene) and vitamin C, to displacement of fat from such diets, or to intakes of other substances in these foods which are not nutrients but may be protective against cancer risk.
   21 CFR 101.78, Health claims: fruits and vegetables and cancer · checked 2026-09-22
7. veg-r5-01 · randomised crossover clinical trial
   The vegetables with the lowest potassium values were spinach (74.85 mg/100 g) and zucchini (93.62 mg/100 g), both cooked under pressure.
   Clin Nutr ESPEN, 2025 · checked 2026-09-22
8. veg-r5-05 · heterogeneity assessment within a meta-analysis
   Consequently, heterogeneity was extremely high (I 2>75 %) for most of the composition parameters for which significant differences were detected.
   Br J Nutr, 2014 · checked 2026-09-22