With this study we present a comprehensive survey of the total antioxidant capacity in foods. Earlier small-scale studies from other laboratories have included from a few up to a few hundred samples [20–22, 29–31], and in 2007 the U.S. Department of Agriculture presented the Oxygen Radical Absorbance Capacity (ORAC) of Selected Foods report including 277 food samples . These studies have been done using different antioxidant assays for measuring antioxidant capacity making it difficult to compare whole lists of foods, products and product categories. Still, a food that has a high total antioxidant capacity using one antioxidant assay will most likely also be high using another assay [20–22]. Consequently, the exact value will be different but the ranking of the products will be mainly the same whichever assay is used. In the present extensive study, the same validated method has been used on all samples, resulting in comparable measures, thus enabling us to present a complete picture of the relative antioxidant potential of the samples.
When classifying the samples into the three main classes the difference in antioxidant content between plant- and animal-based foods become apparent. The results here uncover that the antioxidant content of foods varies several thousand-fold and that antioxidant rich foods originate from the plant kingdom while meat, fish and other foods from the animal kingdom are low in antioxidants. Comparing the mean value of the 'Meat and meat products' category with plant based categories, fruits, nuts, chocolate and berries have from 5 to 33 times higher mean antioxidant content than the mean of meat products. Diets comprised mainly of animal-based foods are thus low in antioxidant content while diets based mainly on a variety of plant-based foods are antioxidant rich, due to the thousands of bioactive antioxidant phytochemicals found in plants which are conserved in many foods and beverages.
Most of the spices and herbs analyzed have particularly high antioxidant contents. Although spices and herbs contribute little weight on the dinner plate, they may still be important contributors to our antioxidant intake, especially in dietary cultures where spices and herbs are used regularly. We interpret the elevated concentration of antioxidants observed in several dried herbs compared to fresh samples, as a normal consequence of the drying process leaving most of the antioxidants intact in the dried end product. This tendency is also seen in some fruits and their dried counterparts. Thus, dried herbs and fruit are potentially excellent sources of antioxidants.
Herbal and traditional plant medicines emerged as many of the highest antioxidant-containing products in our study. We speculate that the high inherent antioxidant property of many plants is an important contributor to the herb's medicinal qualities. In our study we identified Sangre de Grado, the sap from the tree trunk of the species Croton lechleri sampled in Peru to have exceptional high antioxidant content. This sap has a long history of indigenous use in South America for wound healing and as an antifungal, antiseptic, antiviral and antihaemorrhagic medicine. Proanthocyanidins are major constituents of this sap  and studies have shown that Sangre de Grado limits the transcription of a wide range of pro-inflammatory cytokines and mediators and accelerates the healing of stomach ulcers [33, 34] and promotes apoptosis in cancer cells . Other extreme antioxidant rich herbal medicines are Triphala, an Indian Ayurvedic herbal formulation, shown to have anti-inflammatory activity , antibacterial and wound healing properties [37, 38] and cancer chemopreventive potential . Arjuna, another Auyrvedic formula, has been shown to have health beneficial activities [40, 41] while Goshuyu-tou, a traditional Chinese kampo medicine has been shown to significantly reduce the extracellular concentration of NO in the LPS-stimulated Raw 264.7 cells .
With their high content of phytochemicals such as flavonoids, tannins, stilbenoids, phenolic acids and lignans [43–45] berries and berry products are potentially excellent antioxidant sources. The phytochemical content of berries varies with geographical growing condition, and between cultivars [46, 47] explaining the variations found in our study. During the processing of berries to jams, total phenol content is reduced  resulting in lower antioxidant values in processed berry products than in fresh berries.
Nuts are a rich source of many important nutrients and some are also antioxidant-rich. The observed increase in antioxidant content in nuts with pellicle compared to nuts without pellicle is in good agreement with earlier studies showing the flavonoids of many nuts are found in the nut pellicle .
After water, tea and coffee are the two most consumed beverages in the world, although consumption patterns vary between countries. Because of the fairly high content of antioxidants and the frequent use, coffee and tea are important antioxidant sources in many diets. Several different compounds contribute to coffee's antioxidant content, e.g., caffeine, polyphenols, volatile aroma compounds and heterocyclic compounds, [25, 50–52]. Many of these are efficiently absorbed, and plasma antioxidants increase after coffee intake [50, 53]. In green tea, the major flavonoids present are the monomer catechins, epigallocatechin gallate, epigallocatechin, epicatechin gallate and epicatechin. In black tea the polymerized catechins theaflavin and thearubigen predominate in addition to quercetin and flavonols [54, 55].
Interestingly, the antioxidant content in human breast milk is comparable to that in pomegranate juice, strawberries and coffee and on average higher than the antioxidant content observed in the commercially available infant formulas analyzed in our study. Breakfast cereals are also potential important sources of antioxidants; some of these products have antioxidant contents comparable to berries, which are fairly high, compared to other grain products and may be due to antioxidants added to the products in fortification process.
Chocolate have for several years been studied for its possible beneficial health effects . Our results show a high correlation between the cocoa content and the antioxidant content, which is in agreement with earlier studies [30, 57].
As demonstrated in the present study, the variation in the antioxidant values of otherwise comparable products is large. Like the content of any food component, antioxidant values will differ for a wide array of reasons, such as growing conditions, seasonal changes and genetically different cultivars [46, 58], storage conditions [59–61] and differences in manufacturing procedures and processing [62–64]. Differences in unprocessed and processed plant food samples are also seen in our study where processed berry products like jam and syrup have approximately half the antioxidant capacity of fresh berries. On the other hand, processing may also enhance a foods potential as a good antioxidant source by increasing the amount of antioxidants released from the food matrix which otherwise would be less or not at all available for absorption . Processing of tomato is one such example where lycopene from heat-processed tomato sauce is more bioavailable than unprocessed tomato . The large variations in antioxidant capacity observed in the present study emphasize the importance of using a comprehensive antioxidant database combined with a detailed system for food registration in clinical and epidemiological studies.
Initial studies have been carried out to examine the association between intake of antioxidant rich foods and their health effects [67, 70]. Some of these studies describe a beneficial effect on oxidative stress related chronic diseases, e.g. from intake of nuts [49, 69], pomegranates [71–73], tomatoes , coffee , tea [54, 75, 76], red wine [77–79] and cocoa . The highly reactive and bioactive phytochemical antioxidants are postulated to in part explain the protective effect of plant foods. An optimal mixture of different antioxidants with complementary mechanisms of action and different redox potentials is postulated to work in synergistic interactions. Still, it is not likely that all antioxidant-rich foods are good sources and that all antioxidants provided in the diet are bioactive. Bioavailability differs greatly from one phytochemical to another [26, 27, 80], so the most antioxidant rich foods in our diet are not necessarily those leading to the highest concentrations of active metabolites in target tissues. The antioxidants obtained from foods include many different molecular compounds and families with different chemical and biological properties that may affect absorption, transport and excretion, cellular uptake and metabolism, and eventually their effects on oxidative stress in various cellular compartments . Biochemically active phytochemicals found in plant-based foods also have many powerful biological properties which are not necessarily correlated with their antioxidant capacity, including acting as inducers of antioxidant defense mechanisms in vivo or as gene expression modulators. Thus a food low in antioxidant content may have beneficial health effects due to other food components or phytochemicals executing bioactivity through other mechanisms.