**Contents**


## **About the Special Issue Editor**

**Marina Carcea** is a Senior Scientist and Vice-Director of the Research Centre for Food and Nutrition of the Council for Agricultural Research and Economics (CREA) in Rome, Italy—a leading research institution under the aegis of the Italian Ministry of Agricultural, Food and Forestry Policies, where she is also a Scientific Advisor. She has over 30 years of research experience in the fields of agriculture, food, and human nutrition, which she has gained by taking part in, and coordinating, many successful research projects, involving several institutions both within Italy and abroad (Italian Ministries, EU, FAO). Dr. Carcea has authored over 200 scientific publications, many in international journals, as well as several book chapters, and has delivered lectures on her research activity at numerous congresses worldwide over the years. She has been part of multiple national and international committees regarding food and nutrition topics (Italian ministries, technological platforms, UNI, Codex Alimentarius, EU, FAO), and was also a lecturer in food science and technology at the University of Tor Vergata (Rome, Italy) for 12 years. Since 2000, Dr. Carcea has been the Italian national delegate of the International Association for Cereal Science and Technology (ICC) in Vienna (Austria), where she has been actively involved as a member of the executive, governing, and technical committees, and was president from 2011–2012. She is also a member of the ICC Academy and of the Georgofili Academy in Florence (Italy). She is a co-founder of AISTEC, the Italian Association of Cereal Science and Technology, and served as its president from 2009 to 2015. She has been responsible for the organization of numerous successful scientific conferences worldwide, and is a member of the editorial boards of several international scientific journals. Dr. Carcea has been awarded the Harald Perten Prize and the Friedrich Scweitzer Medal for her scientific achievements.

## *Editorial* **Nutritional Value of Grain-Based Foods**

#### **Marina Carcea**

Research Centre for Food and Nutrition, Council for Agricultural Research and Economics (CREA), Via Ardeatina 546, 00178 Rome, Italy; marina.carcea@crea.gov.it; Tel.: +39-06-51494638

Received: 7 April 2020; Accepted: 10 April 2020; Published: 16 April 2020

**Abstract:** Grains are fundamental in the daily diets of many people worldwide; they are used for the production of popular foods, such as bread, bakery products, breakfast cereals, pasta, couscous, bulgur, and snacks. Botanically, they are the seeds of plants, belonging mainly to the groups of cereals, pseudocereals, and legumes. They contribute macronutrients to the human diet, mainly carbohydrates, but also proteins and lipids, and micronutrients, such as vitamins and minerals. They are also an important source of dietary fibre and bioactives, particularly wholegrains, which are of interest for the manufacturing of high value foods with enhanced health benefits. They can be used for the production of gluten-containing (as well as gluten-free) products. One of the main objectives of the food industry when producing grain-based foods is to manufacture safe, attractive products, with enhanced nutritional value to respond to consumer expectations. The following Special Issue "Nutritional Value of Grain-Based Foods" consists of one review and eight original research papers that contribute to the existing knowledge of important ingredients, such as fat substitutes, and of the technological quality and nutritional role of grains and grain-based foods (gluten-containing and gluten-free), such as bread, muffins, and muesli bars.

**Keywords:** cereals; legumes; pseudocereals; gluten-free grains; macronutrients; micronutrients; bioactives; processing; nutrition

Grains are the basis of daily diets for many populations worldwide. They are the seeds of plants, mainly belonging to the botanical groups of cereals, pseudocereals, and legumes.

They contribute macronutrients to the human diet, mainly carbohydrates, but also proteins and lipids, and micronutrients, such as vitamins and minerals. They are also an important source of dietary fibre and bioactives, particularly wholegrains, which are of interest for the production of high value food products with enhanced health benefits [1,2]. Many nutritional guidelines now, in several countries, recommend the inclusion of a greater proportion of wholegrains in the diet for promoting health [3–5]. One of wholegrains roles, recently discovered, refers to their prebiotic activity for gut microbiota, which is fundamental for the host's well-being [6,7]. The content of the aforesaid components varies in grains, depending on genetics and growing conditions, including environment and husbandry.

Humans cannot consume grains in its raw state, so it undergoes a number of processing steps, which might include decortication, dehulling, milling, dough making, extrusion, bread making, couscous making, pasta making, noodle making, bulgur making, etc., up to home cooking [1]. Some grains, thanks to their protein composition, are suitable for the production of gluten-free foods, which are essentially eaten by people suffering from gluten intolerance [8]. Moreover, different kinds of grains can be combined in the same product to take advantage of, in some cases, the complementary composition; thus, producing food with improved nutritional value (see the combination of cereals and legumes that give origin to an excellent aminoacidic composition) [9].

The aim of this special issue was to collect studies on the latest developments in grain science, with regards, in particular, to the improvement of the nutritional value of the raw material due to breeding and/or growing conditions, and the role of processing in keeping or enhancing grains' nutritional potential for the development of healthy, attractive, and improved products (traditional or new) for human consumption.

The contribution of nine papers in this Special Issue, by 12 research groups, from institutions located in six countries, covers a number of topics connected to the nutritional value of grain-based foods, a very important area in food science. Baked food products, bread and muffins in particular, are the object of research in five papers, whereas gluten-free grains/products are covered by two papers; muesli bars and durum wheat grains are also covered by two articles.

Fat provides important sensory properties, such as colour, taste, texture, and odour to baked food products, which often contain high amounts of fat. There is growing demand by consumers for healthier products with reduced fat content, and manufacturers worldwide have started exploring substitution of fats with so-called fat replacers, which range from complex carbohydrates, gums and gels, whole food matrices, and combinations, thereof. The review by Kathryn Colla, Andrew Costanzo, and Shirani Gamlath summarizes the literature on the effect of fat replacers on the quality of baked food products [10]. The ideal fat replacers for different types of low-fat baked products were a combination of polydextrose and guar gum in biscuits at 70% fat replacement, oleogels in cake at 100% fat replacement, and inulin in crackers at 75% fat replacement. The use of oatrim (100% fat replacement), bean puree (75% fat replacement), or green pea puree (75% fat replacement) in biscuits were equally successful.

Excess sodium intake in the diet is associated with high blood pressure and risk of cardiovascular diseases. Bread has been identified as a major contributor to salt intake in the Italian diet; therefore, the research article by Marina Carcea, Valentina Narducci, Valeria Turfani, and Altero Aguzzi presents a survey of sodium chloride (common salt) content in Italian artisanal and industrial bread, to establish a baseline for salt reduction initiatives [11]. Most of the bread consumed in Italy comes from artisanal bakeries; thus, 135 samples of artisanal bread were sampled in 56 locations from Northern to Southern Italy, together with 19 samples of industrial bread representative of the entire Italian production. Salt content between 0.7% and 2.3% g/100 g (as it is basis) was found, with a mean value of 1.5%, Standard Deviation (SD) 0.3. However, the majority of samples (58%) had a content below 1.5%, with 12% having very low salt content (between 0.5 and 1.0%), whereas the remaining 42% had a salt content higher than the mean value, with a very high salt content (>2.0%) recorded for 3% of samples. With regards to industrial bread, an average content of 1.6% was found, SD 0.3. In this group, most of the samples (56%) had a very high content between 2.0 and 2.5%, whereas 5% only had a content between 1.1 and 1.5%.

Bread is also a very versatile product, which, by adequately changing ingredients, can be tailored to cater for the specific needs of some sectors of the population (e.g., the ageing). The research article by Marina Carcea, Valeria Turfani, Valentina Narducci, Alessandra Durazzo, Alberto Finamore, Marianna Roselli, and Rita Rami explores the effects of functional wheat–lentil bread on the immune functions of aged mice [12]. Legumes are considered excellent ingredients to complement cereal composition, so a functional bread, tailored for the needs of the ageing population, was baked by substituting 24% of wheat flour with red lentil flour, and compared with wheat bread. Its nutritional profile was assessed by analysing proteins, amino acids, lipids, soluble and insoluble dietary fibre, resistant starch, total polyphenols, lignans, and antioxidant capacity (Ferric Reducing Antioxidant Power assay). The wheat–lentil bread had 30% more proteins than wheat bread, a more balanced amino acids composition, almost double the minerals as well as total dietary fibre content, double the amount of polyphenols, higher amounts and varieties of lignans, and more than double the antioxidant capacity. The in vivo effect of 60-day bread consumption on the immune response was studied by means of a murine model of elderly mice. Serum cytokines and intraepithelial lymphocyte immunophenotype from the mouse intestines were analysed as markers of systemic and intestinal inflammatory status, respectively. Analysis of immune parameters in intraepithelial lymphocytes showed significant differences between the two types of bread, indicating a positive effect of the wheat–lentil bread on the intestinal immune system, whereas both breads induced a reduction in serum Interleukin-10.

Bread can also be prepared with gluten-free ingredients, such as corn starch and potato starch. The research group by Przemysław Łukasz Kowalczewski, Katarzyna Walkowiak, Łukasz Masewicz,

Olga Bartczak, Jacek Lewandowicz, Piotr Kubiak, and Hanna Maria Baranowska experimented on the substitution of starch with cricket powder as a good source of protein, fat, fibre, and minerals in gluten-free bread [13]. Levels of starch substitutions were 2%, 6%, and 10%; changes caused in the dough rheology and bread texture were studied. While the introduction of cricket powder did not greatly affect dough, the bread was instead characterised by significantly increased hardness and improved consistency. Analyses of water behaviour at the molecular level indicated that cricket powder altered both the bound and bulk water fractions. Moreover, examination of water activity revealed a decreased rate of water transport in samples of bread that contained the cricket powder.

Muffins are also popular bakery products. Generally, they contain high amounts of sugar, and their over-consumption could lead to increased health risks. For this reason, the research group of Jingrong Gao, Xinbo Guo, Margaret A. Brennan, Susan L. Mason, Xin-An Zeng, and Charles S. Brennan studied the potential of modulating reduced sugar (and the potential glycaemic response) in muffins using a combination of Stevia sweetener and cocoa powder [14]. Results illustrate that muffins with 50% replacement of sucrose were similar to the control samples in terms of volume, density, and texture. However, replacement of sugar with 100% Stevia sweetener resulted in reductions in the muffin's height, volume, and increased firmness (by four-fold) compared to the control sample. Sugar replacement significantly reduced the in vitro predictive glycaemic response of muffins (by up to 55% of the control sample).

Grains, together with a variety of other ingredients, such as fruits, nuts, seeds, and chocolate, are also used for the production of so-called muesli bars, generally consumed as snacks. In dietary guidelines across the world, they are often classified as discretionary food due to their (often) high content of fat and added sugars. A comprehensive nutrition review of grain-based muesli bars in Australia, by means of an audit of supermarket products, is provided by the research article by Felicity Curtain and Sara Grafenauer [15]. Their study aimed to provide a nutritional overview of grain-based muesli bars, comparing data from 2019 with those from 2015. Audits of grain-based Muesli bars were conducted in four major supermarkets in metropolitan Sydney, making up more than 80% of total Australian market share. Mean and standard deviation was calculated for all nutrients on-pack, including whole grain per serve and per 100 g. Compared to 2015, mean sugars declined and 31% more bars were wholegrain. Although categorized as discretionary, there were significant nutrient differences across grain-based muesli bars.

Varieties of gluten-free grains are attracting attention as raw materials to improve the nutritional quality of gluten-free foods and to relieve the monotony of a gluten-free diet. In this regard, the research group of Serena Niro, Annacristina D'Agostino, Alessandra Fratianni, Luciano Cinquanta, and Gianfranco Panfili contributed a research article on gluten-free alternative grains: nutritional evaluation and bioactive compounds [16]. The content of thiamine and riboflavin (water- soluble vitamins) as well as that of carotenoids and tocols (liposoluble vitamins) was determined on nine species of cereals and pseudocereals. The analysed samples showed a high content of bioactive compounds: in particular, amaranth, canihua, and quinoa are good sources of vitamin E, while millet, sorghum, and teff are good sources of thiamine. Moreover, millet provides a fair amount of carotenoids, in particular of lutein.

Data about the nutritional composition of gluten-free products are still limited. For this reason, Idoia Larretxi, Itziar Txurruka, Virginia Navarro, Arrate Lasa, María Ángeles Bustamante, María del Pilar Fernández- Gil, Edurne Simón, and Jonatan Miranda determined the composition of gluten-free breakfast cereals, breads, and pasta. They compared the data with equivalent gluten-containing products and were able to produce a research article on micronutrient analysis of gluten-free products. Their low content was not involved in gluten-free diet imbalance in a cohort of celiac children and adolescents [17]. Micronutrient analytical content differences (minerals and vitamins) were observed in gluten-free products when compared with their gluten-containing counterparts. In order to clarify the potential contribution of the gluten-free products to the gluten-free diet's micronutrient shortages, analytical data were used to evaluate gluten-free diets in a cohort of celiac children and adolescents. It does not seem that the lower micronutrient content of the analysed gluten-free products contributed to the micronutrient deficits detected in the gluten-free diets in this cohort (whose diets were not

balanced). Nevertheless, gluten-free products (fortified for folate and biotin) are proposed to prevent the observed deficiencies.

Durum wheat is the raw material of choice for the production of popular foods worldwide, such as pasta, bread, couscous, and bulgur. With the idea of helping officials set proper quality standards for wholegrain durum wheat flours and products where the germ should be preserved, Valentina Narducci, Enrico Finotti, Vincenzo Galli, and Marina Carcea performed analyses and reported in a research article on lipids and fatty acids in Italian durum wheat (*Triticum durum* Desf.) cultivars [18]. The lipids in the durum wheat grain are, in fact, mainly present in the germ. Samples belonging to 10 popular durum wheat cultivars commercially grown in Italy were harvested and analysed for two consecutive years to account for differences due to changes in climatic conditions. Total lipid content ranged from 2.97% to 3.54% dry basis (d.b.) in the year 2010 and from 3.10% to 3.50% d.b. in the year 2011; the average value was 3.22% d.b., considering both years together. Six main fatty acids were detected in all samples in order of decreasing amounts: linoleic (C18:2) > palmitic (C16:0) ≈ oleic (C18:1) > linolenic (C18:3) > stearic (C18:0) > palmitoleic (C16:1). Significant variations in the levels of single acids between two years were observed for three samples.

The above-mentioned nine papers are the result of a variety of original researches performed worldwide on the general topic of grain science; they provide a valuable overview of current issues, which have attracted attention by the scientific community. They represent state-of-the-art research, provide us with updated knowledge, and give us useful indications on the direction of future research on grain science and technology. For these reasons, this special issue "Nutritional Value of Grain-Based Foods" is worth reading, with much attention, by experts in the field, but also by those who just want to know more about this topic.

**Author Contributions:** M.C. planned, wrote and reviewed this article. The author has read and agreed to the published version of the manuscript.

**Funding:** This Editorial received no specific funding.

**Conflicts of Interest:** The author declares no conflict of interest.

#### **References**


© 2020 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Review* **Fat Replacers in Baked Food Products**

#### **Kathryn Colla, Andrew Costanzo and Shirani Gamlath \***

Centre for Advanced Sensory Sciences, School of Exercise and Nutrition Sciences, Deakin University, 1 Gheringhap Street, Geelong 3220, Australia; k.colla@deakin.edu.au (K.C.); andrew.costanzo@deakin.edu.au (A.C.)

**\*** Correspondence: shirani.gamlath@deakin.edu.au; Tel.: +61-3-925-17267

Received: 15 October 2018; Accepted: 22 November 2018; Published: 25 November 2018

**Abstract:** Fat provides important sensory properties to baked food products, such as colour, taste, texture and odour, all of which contribute to overall consumer acceptance. Baked food products, such as crackers, cakes and biscuits, typically contain high amounts of fat. However, there is increasing demand for healthy snack foods with reduced fat content. In order to maintain consumer acceptance whilst simultaneously reducing the total fat content, fat replacers have been employed. There are a number of fat replacers that have been investigated in baked food products, ranging from complex carbohydrates, gums and gels, whole food matrices, and combinations thereof. Fat replacers each have different properties that affect the quality of a food product. In this review, we summarise the literature on the effect of fat replacers on the quality of baked food products. The ideal fat replacers for different types of low-fat baked products were a combination of polydextrose and guar gum in biscuits at 70% fat replacement (FR), oleogels in cake at 100% FR, and inulin in crackers at 75% FR. The use of oatrim (100% FR), bean puree (75% FR) or green pea puree (75% FR) as fat replacers in biscuits were equally successful.

**Keywords:** fat replacers; baked products; carbohydrates; gums; gels; whole foods

#### **1. Introduction**

Dietary fat has an important role within food matrices beyond basic nutrition. It contributes to many sensory and quality properties of a food including physical, textural and olfactory factors which all influence overall palatability. Many snack foods, in particular, rely on dietary fat to fulfil these palatable qualities in order to maintain consumer acceptance and consumption. The World Health Organisation [1], along with many national health authorities [2–5], recommends decreasing consumption of discretionary snack foods due to their poor nutritional content. Excess dietary fat intake, notably from discretionary snack foods, is one of the key contributors to excess energy intake and therefore weight gain [6]. Prevalence of overweight and obesity is rising worldwide [7,8] which is cause for concern as obesity is associated with increased risk of cardiovascular disease [9], type 2 diabetes mellitus [10], and some cancers [11].

Despite consumer awareness and product labelling [12,13], consumption of snack foods is relatively high with little compensation for the increased energy intake [14–16]. Many promoters have been attributed to increased snack food intake, such as convenience, taste, marketing and pricing [16,17]. In order to respond to these recommendations and consumer demands, manufacturing companies are increasingly developing snacks which are more nutrient dense than traditional snacks such as chips and cakes, which are typically high in added fat, sugar and sodium. Some examples of these types of innovative snacks include yoghurts, bars, puddings, crackers and chips which contain popular health foods (or superfoods) such as seeds, nuts, ancient grains, other wholegrains, dietary fibres, legumes, fruits and vegetables. While many of these snacks may be high in protein and dietary fibre, many also typically contribute large amounts of fat, sugar and sodium to the consumers' diet [18].

Efforts must be made to develop appealing snacks which are both high in protein and dietary fibre while not contributing large amounts of sodium, sugar and fat. Snack food categories such as cakes and muffins are yet to see significant innovation in creating high protein or high fibre alternatives [16]. In addition, there are still limited reduced fat options of these baked products on the market, likely attributed to the technological difficulty in producing such products. Ultimately, there is a need to increase the number of nutritious snack options available that satisfy the above drivers, while reducing fat composition and therefore total energy intake. Baked snack foods that omit dietary fat as a "low-fat" alternative often have poor sensory properties, such as crumbliness, dryness, poor mouthfeel and overall reduced consumer acceptance [19–23]. A number of potential "fat replacers" have been purported in order to reduce the fat content in food matrices whilst maintaining the sensory properties that are usually attributed to dietary fat. Fat replacers are subcategorised as either fat substitutes or fat mimetics. Fat substitutes replicate the functional and sensory properties of fat in a food, usually contain no energy or less energy than fat, and may be used to replace some or all of the fat normally present in a product [24,25]. Fat mimetics are protein- or carbohydrate-based ingredients that are not used to fully substitute the use of fat, but rather replicate some of the properties that fat provides within a food [24,25]. Many baked products on the market currently utilise fat replacers in order to reduce the total energy or fat content whilst maintaining consumer acceptance. This review aims to summarise the current evidence for application of fat replacers in biscuits, crackers, muffins, cakes and bread, and their effect on quality and sensory properties.

#### **2. Application of Fat Replacers in Baked Products**

Fat replacers are defined by the American Dietetic Association as "an ingredient that can be used to provide some or all of the functions of fat, yielding fewer calories than fat" [24]. A wide range of products in the food industry uses fat replacers, some of which include meat, dairy and baked products [24]. It is important for product developers and food technologists to understand how different fat replacers influence the sensory and physical quality of snacks in order to guide the development of healthier alternative products. For example, in cakes, fat can contribute to increased leavening, tenderness and a finer crumb through a combined effect of trapping air cells during the creaming process [26]. This structure is then set during baking due to starch gelatinisation and coagulation of egg proteins [26]. Fat is typically used in biscuits to lubricate and coat the flour granules to prevent water absorption, and the development of starch and gluten in order to achieve a fine crumb (crumbly texture) and soft, tender mouthfeel [27]. Fat also contributes other important functions to cakes, biscuits and crackers such as flavour delivery and shelf life which is achieved through delaying water absorption by starch granules [28–31].

Fat replacers can be ingredients which are of carbohydrate, protein or fat origin, with many different types of fat replacers with different structures and functions within each group. We have not differentiated fat substitutes and fat mimetics in this review as the majority of fat replacers used in baked food products are fat mimetics. Instead, we have categorised fat replacers in this review as complex carbohydrate powders, gums and gels, whole food purees and products, or a combination thereof. This categorisation is based on their functional and industrial applications rather than their chemical properties.


#### **3. Summary of the Current Fat Replacers Used in Baked Products**

Complex carbohydrate fat replacers range from digestible starches to non-digestible plant fibres (Table 1). It should be noted that the replacement of dietary fat with complex carbohydrates reduced energy density of all the food products in Table 1, regardless of fibre status, due to complex carbohydrate being less energy dense than fat. The use of fibres instead of starches could have an advantage on the market, as foods may meet criteria for fibre content claims. Inulin, a non-digestible dietary fibre typically derived from chicory root, was observed to have the greatest success in replacing dietary fat in baked products, where a fat replacement (FR) level of up to 75% in legume crackers and cake (1:1, inulin: water; and 1:2 inulin: water, respectively) was able to reduce total energy without any changes in consumer acceptance [33,34]. It should be noted that the addition of inulin did change the textural and physical properties of the cracker and cake products. While acceptance was not measured for the use of inulin in muffins, 50% FR had the least sensory and physical changes compared to 75% and 100% FR [35]. In addition, Zahn et al. tested the use of four commercial inulin formulations in muffins which varied in inulin to water ratio and solubility, but the outcomes for each were similar [35]. Maltodextrin was also successful at 75% fat replacer level in legume crackers and at 66% FR in muffins, although there were changes noted in aroma, appearance, taste and texture [33,36]. Total FR of inulin or maltodextrin (100%) had a significant decrease in consumer acceptance, so it is not recommended to fully replace fat in a baked food product. Results were also promising for inulin used as a fat replacer in biscuits, although there was some notable changes to textural and physical properties [33–35,37–39]. Other complex carbohydrates used as fat replacers in biscuits included lupine extract, maltodextrin, corn fibre, and rice starch, although all of these had significant effects on sensory properties of the biscuits except rice starch [33,36,40–43]. Rice starch has no significant effects on sensory properties, but was only tested at 20% FR. All complex carbohydrate fat replacers had a significant effect on the physical properties of doughs and their baked products, with significant increases in density, toughness, breaking strength, moisture, and decreases in volume for nearly all tested products [33–46].

Of all the complex carbohydrate fat replacers, inulin had the greatest success at reducing total fat and energy of the food product, with the least impact on sensory qualities and consumer acceptance, particularly in the legume crackers. Long chain inulin has the ability to form microcrystals which in turn aggregate together, interact with water, and eventually agglomerate creating a gel network [47]. To some extent, this gel network seems to have the ability to mimic the functions of fat in baked products such as being able to lubricate dry ingredients (through surrounding starch and protein), assisting in maintaining a shortening effect. Maltodextrin was also a successful fat replacer in legume crackers, although it was not as successful at replacing fat in biscuits compared to inulin. Inulin is also a good source of fibre, has promising gut health properties due to its prebiotic nature, and may increase absorption of nutrients such as calcium [48]. Moreover, inulin may benefit from marketing with fibre content claims, which may be appealing to consumers. Therefore, we recommend inulin as a reasonably high level fat replacer in crackers, cakes, biscuits and muffins [48].


**Table 1.** Summary of quality changes of complex carbohydrate fat replacers in baked food products.

aW: Water activity; FR: Fat replacement; NSC: No significant change; ↓: decrease; ↑: increase.

Gum and gel fat replacers, while mostly being carbohydrates, also include lipid-based and protein-based gums and gels (Table 2). Some of these fat replacers may also increase suitability for nutrition content claims, such as sources of fibre or protein. Guar gum and xanthan gum had relatively little effect on the physical properties of the cake product when used as fat replacers [49]. While sensory measures were not compared to a control, both types of cakes were rated as acceptable with a greater acceptance in the cake containing xanthan gum, and 50% FR was considered ideal [49]. Oatrim (a tasteless white powder derived from oats, comprised of amylodextrins and 5–10% β-glucan soluble fiber; incorporated as a powder or gel), caused significant changes to the physical properties of cake, croissants and biscuits [50–52]. However, this did not appear to have any impact on the sensory properties of these foods, even at 100% FR. Pectin also caused significant changes to the physical properties of cake, croissants and biscuits, specifically increasing the hardness and reducing the volume of these foods, which was paralleled by increased perception of hardness and reduced flavour from sensory evaluations [42,46,53,54]. This is notable as pectin was tested at a relatively low FR level in cake and biscuits (10–30%), suggesting it is not an ideal fat replacer in baked products. Hydroxypropyl methylcellulose (HPMC) had significant effects on physical and sensory properties of crackers and biscuits, even at relatively low FR levels [33,55]. While consumer acceptance was not tested on crackers due to being considered unacceptable by a focus group [33], HPMC in biscuits was considered significantly less acceptable compared to control biscuits containing 18% canola oil suggesting it is also not an ideal fat replacer in these foods.

Oleogels are products of solidifying vegetable oils using natural wax esters [56–59]. The oleogelation process forms waxy crystal structure which hold liquid oil within a solid matrix, which allows the use of liquid vegetable oils in place of shortening. While this does not necessarily reduce the total fat content of a food product, it is useful in reducing saturated fat content. It should be noted that all oleogels studies reviewed in this paper did reduce overall fat content of their tested foods [56–59]. However, oleogels were not successful as fat replacers in these studies as they made biscuit and cake denser and harder. Sensory properties seemed to be promising with an increase in taste and no difference in acceptance compared to the control foods. Lastly, whey protein was also not an ideal fat replacer for biscuits as it resulted in a decrease in overall flavour and acceptance [45,46].

Overall, gums and gels were not very successful as fat replacers in baked goods. Oatrim appeared to be the most successful as there were no significant changes to the sensory properties of cake and biscuits, although there were a large range of physical changes to these foods which might have an impact on industrial applications. Xanthan gum and guar gum might potential be useful fat replacers in cake as they had little impact on physical properties, although more robust sensory evaluations are needed in future studies.


**Table 2.** Summary of quality changes of gum and gel fat replacers in baked food products.

aW: Water activity; FR: Fat reduction; NSC: No significant change; ↓: decrease; ↑: increase.

The interest in using whole food fat replacers has increased in recent years. These fat replacers are beneficial as they have a range of carbohydrates, lipids and proteins that may aid in the rheological properties of baked products, making them potentially more suitable than simple extracts and isolates. Overall, whole food fat replacers had the least effect on the physical and sensory properties of baked products, and in some cases increased the consumer acceptance (Table 3). Apricot kernel flour was a successful fat replacer with little impact on the physical and sensory properties of biscuits at a maximum of 50% FR [60,61]. Chia seed mucilage also had little impact on physical properties of cake and bread up to 100% FR [62,63], although sensory properties were not tested in these studies. High oleic sunflower oil (HOSO) did not significantly decrease the amount of total fat in biscuits, but did reduce the saturated fat content [64]. However, the use of HOSO as a fat replacer was not considered successful as it has significant impact on the volume, colour and texture of the biscuits. The use of avocado puree as a fat replacer in cake and biscuits was successful at 50% FR, as it did not impact consumer acceptance [51,65]. However, at 75–100% FR, acceptance of the low-fat cake decreased compared to the control cake containing shortening [65]. Apple puree or pomace was the only whole food fat replacer to result in a reduction in sensory quality and consumer acceptance, even at low FR levels (10%) [66,67]. Therefore, apple puree is not recommended as a fat replacer in biscuits. Bean puree and green pea puree had very similar effects on the sensory properties of biscuits with increases in sensory qualities at 25–75% FR [68,69]. The use of green pea puree at FR of 25% in biscuits was considered ideal, whereas a FR of 100% resulted in reduced consumer acceptance [69]. Lastly, a high β-glucan product derived from oats or oat bran had significant impact on texture, colour and moisture of biscuits [45,54,70]. Although sensory properties were not tested in these studies, this suggests that the high β-glucan product was not a successful replacer for shortening in biscuits.

Whole foods may be the most suitable candidates for fat replacers in baked foods as they appeared to have the least impact on physical and sensory properties. In addition, they may also be beneficial as they may contain phytochemicals and micronutrients which could increase the health benefits and marketing potential of baked foods products, leading to novel functional foods. Lastly, consumer are more likely to accept foods with ingredients or additives that are made from natural, whole food products [71]. Bean and pea purees were the most successful fat replacers for biscuits at 25–75% FR, and avocado puree was successful at reducing fat in cake at 50% FR. However, more studies on whole food fat replacers in biscuits and bread is needed before they can be recommended as reliable fat replacers.


**Table 3.** Summary of quality changes of whole food fat replacers in baked food products.

aW: Water activity; FR: Fat reduction; NSC: No significant change; ↓: decrease; ↑: increase.

*Foods* **2018**, *7*, 192

Fat replacers in combination with additional ingredients may provide better fat-like qualities as the additional ingredients are usually designed to supplement the unwanted effects of individual fat replacers, as seen above (Tables 1–3). These additional ingredients are usually other types of fat replacers, but can also be enzymes or emulsifiers. Few studies have assessed combined fat replacers in baked products, although the results appear promising (Table 4). Polydextrose and guar gum were successful fat replacers in biscuits at a relatively high level of FR (70%), with an increase in perceived taste, flavour and consumer acceptance [72]. Maltodextrin and xanthan gum yielded increased moisture, hardness and chewiness in 66% FR muffins, but sensory analysis was not conducted in these samples [36]. Kel-Lite BK, a commercial fat replacer containing xanthan gum, guar gum, cellulose gel, sodium stearoyl lactylate, gum Arabic, dextrin, lecithin, and mono- and diglyceride, resulted in increased bitterness and, oddly increased both crumb firmness and softness in biscuits at 33%, 66% and 100% FR [54]. HOSO and inulin were also successful fat replacers in biscuits at 100% FR [64,73], although HOSO does contain lipids so the biscuits only had reduced saturated fat rather than total fat. However, HOSO and inulin resulted in decreased appearance, flavour, odour, texture, and consumer acceptance in cakes, croissants and muffins [73]. Therefore, HOSO and inulin may only be suitable for use as fat replacers in biscuits. HOSO and β-Glucan may also be a useful fat replacer at 100% FR as this had little impact on physical properties in biscuits, although sensory evaluations were not conducted [64]. A combination of emulsion filled gel based on inulin and extra virgin olive oil (EVOO) has also been trialed as a fat mimetic in biscuits [74]. At 50% FR, there were no changes to the physical properties and the overall consumer acceptance of the biscuit compared to the control biscuit containing 20% butter, although there was a decrease in overall flavour. However, consumer acceptance was not maintained at 100% FR. Inulin, lipase and a commercial emulsifier ("Colco"; a type of alpha-gel emulsifier containing glycerol monostearate and polyglycerol esters of fatty acids) had little impact on physical properties of cake at 50–70% FR, although no sensory evaluation was conducted for this combined fat replacer either [75]. One study assessed the double, but not triple, combinations of corn fibre, maltodextrin and lupine extract in biscuits, each at 30–40% FR [40]. All combinations had little impact on the physical properties of the biscuits compared to the control biscuit containing 33% margarine. However, consumer preference for corn fibre and lupine extract was significant lower than the control, whereas corn fibre and maltodextrin was significant higher than the control [40]. This suggests that the combination of corn fibre and maltodextrin may be an ideal fat replacement in biscuits at a moderate FR level. Tapioca dextrin, tapioca starch and resistant starch as a combination fat replacer had an impact on a wide range of sensory properties in biscuits [76]. However, overall consumer acceptance decreased, even at relatively low FR levels (10–20%), so we do not recommend the use of this combination fat replacer in biscuits.

Overall, combination fat replacers may be potential candidates for ingredients in low-fat baked products. The use of polydextrose and guar gum appears to be a reasonably effective fat replacer in biscuits. However, with the limited evidence currently available, recommendations cannot be made for the use of combination fat replacers in other baked products.


**Table 4.** Summary of quality changes of combined fat replacers in baked food products.


**Table 4.** *Cont.*

aW: Water activity; FR: Fat reduction; NSC: No significant change; ↓: decrease; ↑: increase, HOSO: High Oleic Sunflower Oil; EVOO: Extra Virgin Olive Oil; EFG: Emulsion Filled Gel.

#### **4. Industry Recommendations and Conclusions**

It should be noted that there is limited literature on the use of fat replacers in low-fat baked products. Many of the reviewed fat replacers have only been assessed once, and also only in one type of food. There is a need for additional replicate studies using a variety of recipes. Also, while we have reviewed the current literature here, we cannot compare physical and sensory properties between studies. Therefore, while we can summarise which fat replacers were successful within a certain baked product, it is difficult to determine which fat replacer is best. In addition, the use of fat replacers in bread, muffins and croissants were only assessed in few studies each. Therefore, there is not enough information to make a recommendation of the best type of fat replacer for these products. Below is our recommendations for the best currently assessed fat replacers in a range of baked food products:

**Biscuit**—Oatrim was the most successful fat replacer in biscuits as it was able to retain most sensory properties of a traditional biscuit even at 100% FR, although there was a decrease in hardness and brittleness [51,52]. However, it should also be noted that both bean puree and green pea puree were able to increase the sensory qualities and consumer acceptance of biscuits at 75% FR with less of an impact on the physical properties compared to oatrim [68,69]. Legume purees might also have an advantage over oatrim as they may aid the marketability of food products due to potential nutrition claims such as vegetable and protein content. However, legume purees should not be used at 100% FR. Overall, we recommend the use of either oatrim or legume purees as fat replacers in biscuits.

**Cake**—Oleogels appeared to be the most successful fat replacer in cake, with no changes to the sensory qualities at 100% FR [57–59]. However, there were significant changes to the physical properties of cake when using oleogels at FR levels ≥50% [58] which might lead to difficulty during cake production. An alternative could be avocado puree which was only successful at 50% FR but had less of an impact on the physical properties of cake [65], or inulin which was successful up to 75% FR but had an impact on the physical and textural properties of cake [34].

**Cracker**—While there was only one study on the use of fat replacers in crackers [33], it assessed and compared a range of fat replacers in the one study. Inulin appeared to be the most successful fat replacer in these crackers, reaching an acceptable level of FR at 75%. The additional benefits of using inulin is that it may aid the marketability of food products due to potential high fibre claims.

**Author Contributions:** K.C. and A.C. collated the literature and wrote the manuscript. S.G. conceived the original idea, designed the format, reviewed the manuscript and provided critical feedback.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **A Survey of Sodium Chloride Content in Italian Artisanal and Industrial Bread**

#### **Marina Carcea \*, Valentina Narducci, Valeria Turfani and Altero Aguzzi**

Research Centre for Food and Nutrition, Council for Agricultural Research and Economics (CREA-AN), Via Ardeatina 546, 00178 Roma, Italy; valentina.narducci@crea.gov.it (V.N.); valeria.turfani@crea.gov.it (V.T.); altero.aguzzi@crea.gov.it (A.A.)

**\*** Correspondence: marina.carcea@crea.gov.it; Tel.: +39-(0)6-51-494-638

Received: 19 October 2018; Accepted: 2 November 2018; Published: 5 November 2018

**Abstract:** A nationwide survey on salt content in both artisanal and industrial bread was undertaken in Italy to establish a baseline for salt reduction initiatives. Excess sodium intake in the diet is associated with high blood pressure and the risk of cardiovascular diseases. Bread has been identified as a major contributor to salt intake in the Italian diet. Most of the bread consumed in Italy comes from artisanal bakeries so 135 artisanal bread were sampled in 56 locations from Northern to Southern Italy together with 19 samples of industrial bread representative of the entire Italian production. Sodium chloride content was analysed according to the Volhardt's method. A salt content between 0.7% and 2.3% g/100 g (as is basis) was found, with a mean value of 1.5% (Standard Deviation, 0.3). However, the majority of samples (58%) had a content below 1.5%, with 12% having a very low salt content (between 0.5% and 1.0%), whereas the remaining 42% had a salt content higher than the mean value with a very high salt content (>2.0%) recorded for 3% of samples. As regards the industrial bread, an average content of 1.6% was found (SD, 0.3). In this group, most of the samples (56%) had a very high content between 2.0% and 2.5%, whereas 5% only had a content between 1.1% and 1.5%. Statistics on salt content are also reported for the different categories of bread.

**Keywords:** salt; sodium chloride; artisanal bread; industrial bread

#### **1. Introduction**

One third of global deaths are due to cardiovascular diseases, including heart attacks, strokes and related diseases (World Health Organization, 2007). High blood pressure is the major risk factor and, according to a substantial body of epidemiological and interventional studies, an excess of sodium in the diet is the primary cause of hypertension [1–5]. Salt intake is thus being increasingly monitored and evaluated worldwide. The human physiological need of sodium is rated around 130–230 mg/day by the World Health Organization (WHO), but in many industrialized countries sodium intake is actually 3600–4800 mg/day [6]. This indicates that the mean salt intake of populations is well in excess of dietary needs and far from the WHO recommendation to have a salt intake <5 g/day [6], that is, 2000 mg/day of sodium.

In the last decades, a wide range of initiatives aimed at salt reduction (DASH: Dietary Approaches to Stop Hypertension, WASH: World Action on Salt and Health, National Salt Reduction Weeks, CASH: Consensus Action on Salt and Health) have been started at the international level to sensitize people about salt consumption and salt content in some food categories, to educate the population about the dangers of salt in excess, and to translate scientific evidence into public health policies and plans for reformulation of processed foods [1,3,5,7–11]. In fact, processed foods are the main source of salt in the diet, with cereal products contributing the most of the overall intake [6,10,12,13], especially in those countries where bread is consumed daily at every meal. A recent survey highlighted

an average yearly consumption per capita of 64 kg in Europe with Italy ranking third after Germany and France (57 kg) [14].

When in 2008 the European Commission (EC) launched the EU Framework for National Salt Initiatives, an interdisciplinary Working Group for reduction of salt intake (GIRCSI) was established in Italy at the Ministry of Health [3] with the main objective to device strategies to reduce salt consumption in the population. Bread was identified as one of the first processed foods to address and the first steps to be taken were to measure and monitor the sodium content of bread to promote reformulation of foods containing less salt. Other European countries have launched initiatives to reduce salt content in bread and recently news has appeared on the Internet that Portugal will set mandatory maximum salt levels in bread by 2019 [15].

This paper represents the first comprehensive survey on the salt content in bread consumed by the Italian population, and the data reported here represent the baseline for the reformulation of salt reduced bread. Most of the bread consumed in Italy is produced by artisans in artisanal bakeries according to different recipes and procedures and only a small proportion of the market (around 10%) is covered by the industrial production: consequently, a great variability in salt content was expected. Several breads in Italy are also protected by European authenticity labels such as Protected Designation of Origin (PDO) and Protected Geographical Indication (PGI) labels. Both artisanal and industrial bread was considered in the present study. Moreover, a comparison between methods to determine Na content in flour and bread was made on selected bread samples to assess the reliability of the quick method which was used for sodium chloride determination in bread.

#### **2. Materials and Methods**

#### *2.1. Samples and Sampling Method*

Artisanal bread was purchased at selected bakeries in Northern, Central and Southern Italy particularly in places with a specific identity in terms of bread production. In each bakery, the most consumed types of bread were sampled. For the industrial sector, samples of all the Italian production available on the market were purchased at supermarkets and included sliced pan bread (12 samples) and "traditional-like" bread (6 samples). In total, 154 bread samples (kinds of bread) were collected, between winter 2009 and spring 2010. For each type of sample, a spreadsheet was filled with data concerning origin, ingredients, weight and baking method.

In detail, 19 samples of industrial bread were collected together with 135 samples of artisanal bread from 56 locations (Figure 1). Seven out of 154 samples (1 sample of industrial bread and 6 samples of artisanal bread) were declared, at purchase, without salt and subsequent analysis performed by us, confirmed this feature.

Samples of baking wheat flour (*Triticum aestivum* L. flour, which is the kind of flour mostly used in bread baking in Italy) of two different extraction rates according to the Italian law (0 and 00, ash content maximum 0.65% and 0.55% on dry matter, respectively) were purchased at a local supermarket and analysed for their sodium content.

#### *2.2. Analytical Methods*

Soon after purchase, representative portions of each type of bread were cut in small pieces, well homogenised and used for the following analyses. A portion of the sample was used to determine moisture according to ICC Standard No. 110/1 [16], whereas another portion was prepared according to AACC method 62-05 "Preparation of sample: bread" [17] by drying it at 35 ◦C overnight and grinding it by a MLI 204 laboratory mill (Bühler, Uzwil, Switzerland). The residual moisture in the sample was also determined according to the previous ICC Standard. The determination of chloride ion in bread samples was carried out by titration according to the AACC method 40-33 "Chloride in yeast foods—quantitative method (Volhardt's method)" [17]. Sodium chloride content was finally calculated based on the content of chloride ions in sample. Duplicate analysis was carried out for each

sample. Duplicates differing by more than 0.20 were rejected and analysis repeated. Salt content in bread was expressed as percentage, as is basis.

**Figure 1.** Bread samples collected in the different regions of Northern, Central and Southern Italy. (**a**) Number of samples from each Italian region. Northern regions are coloured in dark grey, Central regions in white and Southern regions in light grey (division according to the Italian Central Institute of Statistics). (**b**) Bread samples of different size, shape and ingredients.

A selection of bread and wheat flour samples were also analysed by Inductively Coupled Plasma Spectroscopy (ICP) on a Perkin-Elmer Plasma Optima 3200XL (Perkin-Elmer, Waltham, MA, USA) in order to determine sodium content in the raw material, and confirm that the results obtained by the AACC method were in good match with those obtained by ICP. Samples were first mineralized in nitric acid (6 mL HNO3 + 1 mL H2O2) in a microwave oven (Milestone 1200 Mega, FKV srl, Torre Boldone, Italy). Standard CRM 189 (whole meal flour) from the Community Bureau of Reference (BCR, Brussels, Belgium) was used as a Reference Material.

#### *2.3. Statistics*

The seven samples of bread without salt were excluded from statistical elaboration. Statistical determination of mean, standard deviation and percentage distribution were performed using Microsoft Office Excel 2007. For easiness of results understanding and interpretation, it was decided to establish 4 classes of salt content (as is basis): (i) 0.5–1.0% (low salt content); (ii) 1.1–1.5% (medium salt content); (iii) 1.6–2.0% (high salt content); and (iv) 2.1–2.5% (very high salt content).

The percentage distribution in the above-mentioned salt content classes was calculated for 14 groups that represented all the different commercial categories that could be found in our sample population: all samples together, industrial vs. artisanal samples, 4 categories according to weight, 5 categories according to ingredients, and 2 categories according to leavening method.

#### **3. Results**

This section presents the results of analyses of sodium content in soft wheat white flour widely used for bread baking in Italy, and eight samples, selected for their different characteristics and presumably different salt content, are reported in Table 1. The same table briefly describes each sample compositional or processing characteristics. One column reports data obtained by calculating the sodium content in samples analysed by the standard AACC method 40-33 (Volhardt's method) [17], whereas the other column refers to the sodium content in samples determined by means of ICP.

The purpose of this study was to assess the contribution of the raw material flour to the salt content in bread, to verify whether the bread declared to be without any salt actually had a negligible sodium content, and whether the data obtained by the Volhardt's method could be compared with those obtained by a more sensitive but more complex and expensive method.

Data reported in Table 1 show that sodium was not detected in both types of commercial soft wheat white flours, even in the 0 type which is less refined than 00. Based on this result obtained with a very sensitive instrument, it was decided not to analyse these two samples by the Volhardt's method.

No sodium was detected following both analytical procedures in the three different bread samples declared by the bakers to be without salt addition. Sodium was detected by means of both methods in the five remaining samples and values ranged 03–06 for the Volhardt's method and 0.1970–0.4902 g/100 g (as is sample) for the ICP method. In both cases, the highest value was obtained for durum wheat bread.


**Table 1.** Sodium content in flour and bread samples as measured by two methods of different sensitivity.

\* Average of two determinations on as is sample.

The statistical elaboration of salt content data referring to the 147 samples of salty bread is reported in Figures 2–5. In our survey, a salt content in bread ranging between 0.7% and 2.3% (as is basis) was found, with a mean value of 1.5% and a standard deviation (SD) of 0.3 (Figure 3). If we look at the distribution of salt content in the different classes as specified in the Materials and Methods Section, we can see that the majority of bread samples (58%) had a salt content below the reported mean value (>1.5%) (Figure 2a) with 12% having a very low salt content falling within the range 0.5–1.0%, whereas the remaining 42% had a salt content higher than the mean value with a very high salt content (>2.0%) recorded for 3% of samples.

**Figure 2.** Percent distribution of bread samples according to salt content classes.

**Figure 3.** Percent distribution of industrial bread samples according to salt content classes.

If we have a separate look at the artisanal and the industrial production (Figure 2b,c), we can say that, although the average salt content in bread is very similar (1.5 and 1.6 g/100 g as is basis with standard deviations of 1.1 and 0.3, respectively), the distribution of our samples in the different salt content classes is different. In the artisanal bread, the majority of bread samples (61%) had a salt content below the reported mean value (>1.5%) (Figure 2a) with 14% having a very low salt content falling within the range 0.5–1.0%, whereas the remaining 39% had a salt content higher than the mean value with a very high salt content (>2.0%) recorded for only 3% of samples. In the industrial production, only three classes were represented, the very low salt content class (<1.0 g/100 g as is) having disappeared. Most of the samples (56%) had a very high salt content between 2.0 and 2.5 g/100 g (as is) whereas only 5% had a salt content between 1.1 and 1.5 g/100 g (as is).

A further differentiation can be made within the industrial bread by considering separately the sliced pan bread, which represents the most consumed category, and the so-called "traditional-like" bread which resembles more in its shape the artisanal bread (Figure 3). In the pan bread, a mean value of 1.5 g/100 g (as is) was obtained (SD 0.3) and two salt content classes (1.1–2.0%) were found, each having a 50% share, whereas in the traditional-like bread an average value of 1.8% g/100 g (as is) was found (SD 0.3) which derived from the contribution of three salt content classes (1.1–2.5%) with the very high salt content class having a share of 16.5%.

**Figure 4.** Percent distribution of artisanal bread samples of different weight according to salt content classes.

**Figure 5.** Per cent distribution of artisanal bread samples, differing in dough formulation and leavening method, according to salt content classes.

Given the great variety of artisanal bread, we thought it would be interesting to compare the salt content in different types of bread to determine whether there was any relationship between specific bread characteristics and salt content: weight, ingredients and leavening method were identified as interesting quality traits. The 129 artisanal bread loaves were, therefore, grouped into three different categories according to their weight, ingredients and leavening method. Within the "weight" category, four classes were identified based also on the bread shape: (i) 25–95 g (48 samples); (ii) 100–250 g (37 samples); (iii) 300–700 g (27 samples); and (iv) 1000–2000 g (17 samples), with rolls, typical of the bread production in Northern Italian regions, and big loaves typical of Central and Southern regions. Four classes were also established in the "ingredients" category as follows: common white bread, whose dough is typically formulated with just soft wheat flour, water and salt (66 samples); brown bread, with different amounts of soft wheat whole-meal flour in addition to the common white bread ingredients (24 samples); durum wheat bread (20 samples), typical of Southern Italy but also appreciated and consumed all over Italy made with remilled durum wheat semolina, water and salt; and "special" bread, that is, soft wheat white bread with other ingredients such

as oil, milk, and potatoes (19 samples). As regards the leavening method, two classes were established: sourdough and compressed yeast.

Figure 4 reports the pie charts of the percentage distribution in the four salt content classes according to the weight of the bread. The most represented weight class was small breads, i.e., rolls (48 samples), and the least represented was big loaves weighing up to 2 kg. This distribution actually reflects the pattern of consumption of the Italian population. The categories up to 250 g were the most represented. Although the average salt content and SD is very similar or identical in the four groups and goes from 1.4 to 1.5 g/100 g (as is), (SD, 03 and 0.5, respectively), the percentage distribution of the four salt content classes was different and peculiar within each group with the highest salt content class not being represented for example in the smallest bread group and the biggest loaves having the highest percentage of samples (6%) having a salt content between 2.0% and 2.5% (as is).

For dough formulation (Figure 5), we obtained a mean value of 1.4 g/100 g (as is) (SD, 0.4), for brown bread, 1.5 g/100 g (as is) (SD, 0.4 and 0.2, respectively, for common bread and special bread), and for durum wheat bread, 1.6 g/100 g (as is) (SD, 0.3). In the durum wheat group, only two salt classes were found, namely 1.1–1.5% and 1.6–2.0%, with the first being more represented (61%) than the latter (39%).

The two leavening methods had very different sizes, with sourdough samples being only 21 while compressed yeast bread samples being 108. These numbers actually reflect the presence of these categories on the market with sourdough bread being less frequently found. However, the two groups had the same average salt content, 1.5 g/100 g (as is) (with SD = 0.3 for sourdough bread, and SD = 0.2 for compressed yeast bread). In the sourdough bread group, there were no samples with a very high salt content (≥2.1%).

#### **4. Discussion**

Recently, several similar surveys have been conducted in countries where bread is a staple food and has therefore been identified as a major contributor to the daily intake of salt and sodium in the population [18–21].

In our study, the analysis of sodium content in a selection of commercial refined wheat flour and bread samples by ICP analysis showed that salt content in white bread, which is the most consumed type of bread in Italy, is not due to a natural occurrence of sodium in the flour, but to the salt added in the recipe. The higher sensitivity of the ICP analysis than the Volhardt's method enabled to confirm, in fact, that sodium naturally occurring in the white flour is negligible (Table 1) and, moreover, it showed that salt content in some of the sampled bread samples, declared at purchase to be "without salt", was, in fact, below 0.1%.

Even if there is no perfect correspondence between the results obtained by the two methods (Table 1), it is nevertheless interesting to notice that the ranking of the samples as regards their sodium content was the same. These results confirmed the practical value of the Volhardt's method for the determination of sodium chloride in bread and for the purpose of our study.

Although the average salt content found in all our bread samples (1.5% g/100 g, as is basis) is similar to that reported in the literature for other European countries [22], the range of values found was very wide with the highest values around 2.3%. This means that there is room for improvement and that salt reduction initiatives and campaigns are advisable also in Italy.

The statistical elaboration of data also showed an interesting variation of salt content in bread at geographical level. It emerged that the mean salt content in bread produced and consumed in Central Italy is slightly lower than in the north and south of the country. In fact, the mean salt content was 1.3% in the 52 bread samples from Central Italy with a SD of 0.4, whereas it was 1.6% with a SD of 0.2 in the 38 bread samples from Northern Italy, and 1.5% with a SD of 0.3 in the 39 bread samples from Southern Italy. In detail, it emerged that in Northern Italy there is no share of bread with a salt content below 1.0%, whereas 21% of analysed samples purchased in Central Italy and 15% of bread types sampled in Southern Italy were in this range. These figures confirm the existence of a well-established tradition in some regions of central Italy, e.g., Umbria, Marche and Tuscany, of producing bread loaves with a very low or null salt content. This evidence also hints at the fact that the main problem in salt reduction might be consumers' acceptance and salt content in bread might be reduced at the artisanal level without encountering too many technological problems.

Considering separately the artisanal production from the industrial production, even though in Italy the latter represents one fourth of the former, it is interesting to notice that the average salt content is higher in industrial bread (1.6% g/100 g, as is basis, with a SD of 0.3) than in the artisanal bread (1.5% g/100 g, as is, and a wider SD 1.1). and no samples were found falling within the class containing a small amount of salt (0.5–1.0%). Most industrial samples (56%) fall in the high salt content class (1.6–2.0%), whereas artisanal bread's most represented category (47%) is that of 1.1–1.5% salt content (medium salt content). The industrial production can easily be subdivided into two categories, namely pan bread (which is always sliced) and traditional-like bread which is more similar in shape and appearance to artisanal bread. They represent the two most common types of industrial bread that are produced by a few manufacturers in a homogeneous and standardized way, and distributed all over the national territory. It is interesting to notice that the pan bread had a more homogeneous salt content, ranging from 1.1% to 2.0% with an average of 1.5%, as is, and a SD of 0.3, whereas the traditional-like bread had 16.5% of samples having a salt content between 2.1% and 2.5% and a higher average content of 1.8% and the same value (0.3) of standard deviation.

The average content in Italian industrial bread is higher than that reported in other European countries such as UK, where in 2011 a National survey, promoted by the Consensus Action on Salt and Health (CASH), reported for industrial pre-packaged bread a salt content ranging between 0.58% and 0.83% [7].

In addition, in the industrial Italian production, it is advisable to reduce the salt content and, considering that most of the production is in the hands of few manufacturers, it should not be too difficult to reach this target. Moreover, being industrial bread generally supplied to canteens, hospitals and caterings, there are high chances that salt reduction initiatives can reach a broad number of consumers in a very short time even if the artisanal market share represents the biggest challenge for any future salt reduction initiative.

The analysis of salt content in bread according to its weight showed two significant pieces of evidence. In big loaves weighing 1000–2000 g (Figure 4d), there is a more consistent percentage of samples (35%) with a very low salt content (0.5–1.0%, as is basis). On the other hand, rolls weighing 25–95 g (Figure 4a) proved to be the only weight class with a salt content always below 2.0% and never reaching the very high content. Comparing the results obtained for the four classes under consideration with the mean salt content obtained for artisanal bread (1.5%, as is basis, with SD of 1.1), it emerged that a very good share of samples for each class has values below this mean: 57% of rolls (class 25–95 g), 54% of small loaves (class 100–250 g), 66% of medium loaves (class 300–700 g) and 70% of big loaves (class 1000–2000 g).

Considering dough formulation, i.e., the different raw materials used in bread making (Figure 5), it emerged that durum wheat bread had a more homogeneous salt content than common, brown or special bread: all samples belonged to only two salt classes, namely 1.1–1.5% and 1.6–2.0%. The main share (61%) is due to the lower salt content class. Considering the mean value of salt content in artisanal bread as a reference point for discussion, it was observed that 59% of common bread (Figure 5a), 63% of brown bread (Figure 5b), 61% of durum wheat bread (Figure 5c) and 53% of special bread (Figure 5d) samples have a salt content lower than this mean.

Bread samples with a salt content exceeding 2% belonged only to the class "common bread" and "brown bread", but at the same time brown bread is the category with the highest percentage (27%) of samples with a very low salt content (0.5–1.0%, as is basis) followed by common bread (15% of samples). The main difference between the sourdough and compressed yeast bread categories can be seen in the presence of 4% samples with a very high salt content (2.1–2.5%). The average content

is the same for both categories, i.e., 1.5%, as is, but the SD is higher (0.3 versus 0.2) for sourdough bread. By focusing on the results obtained for the sourdough bread and brown bread categories, which had a significant percentage of the very low salt content, it could be speculated that the use of the sourdough and the formulation with wholemeal flours, can add to bread a natural flavour that prevents an excessive addition of salt to the dough.

#### **5. Conclusions**

The present study represents the first extensive survey on the actual salt content in Italian bread and provides the baseline for national salt reduction initiatives, as recommended by the European Commission (EC) to each country within the EU Salt Reduction Framework [8].

As regards artisanal bread, which is the type of bread mostly consumed by the Italian population, the survey highlighted a great variability of values obtained for salt content (from 0.7% to 2.3%, as is basis) that enabled both the identification of a market share offering bread with a high-salt content (2.0–2.5%) that should be immediately addressed by salt reduction policies and education campaigns, as well as the existence of a substantial share of bread with a low salt content that is in line with the EC and WHO recommendations. A good share of the Italian bakery market is represented by the long-established tradition of bread produced with a low salt content (0.5–1.0%) and widely consumed in some regions of Central Italy, e.g., Marche, Toscana and Umbria. This evidence indicates that technological strategies for low-salt bread manufacturing and campaigns for consumer education to gradual salt reduction in bread are possible with high chances of success.

As regards industrial bread, there is less variation in salt content compared to artisanal bread but it is on the high content side. However, future initiatives for salt reduction are more likely to be successful and reach in shorter times a major share of consumers because industrial bread production is controlled by a few manufacturers that distribute their standardized products all over Italy.

**Author Contributions:** M.C. planned and led the research and wrote the paper. M.C., V.N. and V.T. sampled the bread. V.N. and V.T. performed the experiments helped by A.A. who performed the ICP analysis. V.N. performed the statistical calculations and produced the figures.

**Funding:** This work was carried out within the framework of the MINISAL project funded by the Italian Ministry of Health.

**Acknowledgments:** The authors wish to thank Vittorio Vivanti (INRAN, Italy) and Licia Iacoviello (University of Molise, Italy) for their valuable collaboration in bread sampling, Paolo Fantauzzi for his technical assistance and Francesca Melini for her bibliographical help.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

#### *Article*
