ABSTRACT

Background: There is a lack of evidence-based resources available to guide the formulation of nutritionally adequate low-carbohydrate (LC) diets.

Methods: We searched the FSANZ Australian Food Composition Database to identify the top 10 LC food sources for each of the 27 essential micronutrients (vitamins, minerals, and fatty acids). To be included, foods had to contain ≤5.0g total digestible carbohydrates per serve (serving), contain at least 20% of the highest available Nutrient Reference Value (NRV) per serve, and be classified as a minimally processed food. We then categorized the foods into practical food groups and developed a sample meal plan suitable for a very LC diet (<30g/day) using these resources.

Results: Twelve (12/27) nutrients had 10 LC foods containing ≥20% NRV of that nutrient per serve for both men and women. Nine (9/27) nutrients had 10 LC foods containing ≥20% NRV for women but not for men, and six (6/27) nutrients had <10LC food sources for both women and men. Vitamin A, vitamin B12 and zinc had entirely animal food sources, and 16/27 nutrients had >80% animal food sources. Vitamin C had entirely plant food sources, and two additional nutrients (vitamin B7 and magnesium) had more than 80% plant food sources. The top 10 LC foods across all 27 nutrients were categorised into 18LC food groups. The group representing the highest number of nutrients was oily fish followed by seafood, offal, white fish, nuts, legumes, seeds, eggs, poultry, non-cruciferous vegetables, pork, ruminant animals, game meat, cruciferous vegetables, dairy, fruit, fungi, then grains. The sample meal plan met or exceeded the highest NRVs by ≥90% for 24/27 and 25/27 nutrients, for men and women respectively.

Conclusion: This study presents practical information to support healthcare practitioners and future researchers in formulating nutritionally adequate LC dietary plans.

Keywords: Diet Therapy; Diet; Carbohydrate- Restricted; Micronutrients; Nutrients; Nutritionists

Abbreviations: LC: Low-Carbohydrate; VLC: Very Low-Carbohydrate; ADGs: Australian Dietary Guidelines; AGHE: Australian Guide to Healthy Eating; TEI: Total Energy Intake; NRVs: Nutrient Reference Values; RDI: Recommended Dietary Intakes; AI: Adequate Intakes; FSANZ: Food Standards Australia New Zealand; SDT: Suggested Dietary Target

Introduction

Low-carbohydrate (LC) diets include dietary approaches containing less than 130 grams (g) of digestible carbohydrates daily, with very low-carbohydrate (VLC) diets typically containing less than 50 g/day. LC diets have become increasingly popular over recent years, and their health benefits have been demonstrated in numerous clinical populations, including type 2 diabetes [1,2], polycystic ovarian syndrome (PCOS) [3-5], type 1 diabetes [6,7], the metabolic syndrome [8,9], overweight and obesity [10], and epilepsy [11-13]. However, a 2019 systematic review analysed changes in micronutrient intakes for various LC diets investigated in clinical trials and found that intakes of certain nutrients decreased significantly, raising concern about the potential increased risk of nutritional deficiencies [14].

LC diets are not currently included as a recommended dietary pattern in national dietary guidelines, including the Australian Dietary Guidelines (ADGs) [15,16]. The Australian Guide to Healthy Eating (AGHE) [17] is a nationally recognised resource developed from the ADGs [16] to guide the formulation of high-carbohydrate diets (45-65% total energy intake [TEI]) using specific proportions of various food groups to meet the Nutrient Reference Values (NRVs) for key nutrients [18]. It encourages Australians to base their diets on grain and cereal products that have been fortified with key nutrients, such as thiamine, folic acid and iodine [15]. Adults aged 19-50 years are recommended to consume six daily serves of bread, rice, pasta, porridge, cereal, muesli, crumpets and scones [16]. Given that grain and cereal products are high in carbohydrates (starches and/or sugars), they cannot be included in LC diets in the quantities recommended by national authorities. This presents a practical challenge for researchers and healthcare practitioners attempting to formulate LC dietary plans that meet the NRVs, and there is an urgent need for evidence-based information and practical resources on this topic to minimize the risk of nutritional deficiencies.

Previous research demonstrates that LC diets can in fact meet nutritional requirements for adults [19] and adolescents [20], despite macronutrient proportions not aligning with national dietary guidelines. Zinn et al. 2018 showed that two LC meal plans (61-69g/day of carbohydrates) exceeded NRVs, including Recommended Dietary Intakes (RDI) and Adequate Intakes (AI), for all nutrients except iron for females [19]. However, the study provided no practical guidance on how to construct nutritionally adequate LC dietary plans. Similarly, Jebeile et al. 2020 assessed the nutritional adequacy of different VLC and LC meal plans with <30g, <50g and <130g/day of carbohydrates [20]. The results suggest that it is particularly challenging to meet nutritional requirements on a VLC diet. The <30g/day meal plan was below the RDIs for calcium, iodine and magnesium, in addition to being below the female RDI for iron [20]. The authors concluded that to maximize nutrient values, carbohydrates should be increased to 120g/day [20]. There appears to be a gap in the knowledge and understanding of how to source essential nutrients when dietary carbohydrates are restricted.

To our knowledge, no study has taken a nutrient-centered approach to identify specific foods and/or food groups that can assist researchers and healthcare practitioners in formulating LC dietary plans to be nutritionally adequate. Therefore, the aim of this study was to present practical information on meeting the NRVs for essential vitamins, minerals, and fatty acids using a LC dietary approach. Specific objectives of this study were to identify the top 10 LC food sources for each of the essential vitamins, minerals, and fatty acids, and categorise these foods into practical LC food groups. We then used these resources to develop a sample meal plan suitable for a VLC diet.

Materials and Methods

Data Collection

We searched the Food Standards Australia New Zealand (FSANZ) Australian Food Composition Database [21] between March 2020 and February 2021 to identify the top 10 LC food sources for 27 essential micronutrients. The nutrients assessed were vitamin A, vitamin D, vitamin E, thiamine (vitamin B1), riboflavin (vitamin B2), niacin (vitamin B3), pantothenic acid (vitamin B5), vitamin B6, biotin (vitamin B7), folate (vitamin B9), vitamin B12, vitamin C, calcium, iodine, chromium, magnesium, iron, fluoride, sodium, potassium, selenium, copper, zinc, molybdenum, manganese, phosphorous, and total long-chain omega-3 fats. Where data for multiple forms of a single nutrient were available (e.g., natural folate and folic acid), we used the preformed or most biologically active versions of the nutrients for analysis, which included: folate (natural), niacin (preformed B3), vitamin D3 equivalents, retinol (vitamin A), vitamin E (total vitamin E activity of a food) and total long-chain omega-3 fatty acids (DHA + EPA + DPA). Choline and vitamin K are essential nutrients that were not available in the Australian Food Composition Database [21], so these nutrients could not be assessed. In addition, linoleic acid (total omega-6 fats) could not be assessed because the database only presented data according to each individual omega-6 fatty acid, and it was not possible for us to analyse this data in the same way as the other nutrients.

Data Synthesis and Analysis

i) Identifying the Top LC Sources of Essential Micronutrients

The Australian Food Composition Database (searchable online database) [21] was searched according to each essential nutrient. For each nutrient, foods were ranked from highest to lowest (nutrient content per 100g) and individually assessed, in order, against pre-specified eligibility criteria by two independent researchers (JT, RF or NS). For each food, the following data items were extracted from the database for assessment: food description, nutrient content per 100g, nutrient content per serve (serving), and total (digestible) carbohydrates per serve. To be included in the top 10 food list for a nutrient, foods had to: contain ≤5.0g of total digestible carbohydrates per serve, ≥20% of the NRV per serve (based on meeting 100% of the NRV with five eating occasions daily), and be classified as whole or minimally processed food according to the NOVA classification system (NOVA 1-3) (Supplementary Table 1: NOVA Classifications [22]). The database was searched until 10 different foods for each nutrient were deemed eligible and included. Two independent researchers compared their results for any discrepancies (JT, RF or NS) and if consensus could not be reached, a third researcher (KR) made the decision.

Food serve sizes were taken from the ADGs, where possible [16]. For foods that were not specified in the ADGs, a registered dietitian (JT) designated a serve size based on average serve sizes provided by food manufacturers and/or what was considered to be a practical amount of the food (Supplementary Table 2). The highest available NRVs for adults aged 19-50 years were used to assess the eligibility of foods. In order of preference, we used the Suggested Dietary Target (SDT), RDI or AI [18]. If there were ≤10 foods containing ≥20% of the highest NRV per serve for a nutrient (e.g., potassium), the database search continued until 10 different foods were identified and the %NRV values of all included foods were reported in the results for transparency. Foods were excluded from the top 10 list for a nutrient if they: contained >5.0g total digestible carbohydrates per serve, were classified as NOVA 4 foods (ultra-processed foods) [22], were herbs, spices, salts, or leavening agents (e.g., yeast or baking powder), or were synthetically fortified with the nutrient of interest (e.g., tofu fortified with B12). Additional coding items describing how we dealt with similar foods, raw and cooked foods, different foods with the same nutrient value, and foods with non-edible components are included in the Supplementary Files (Supplementary Table 3), as well as a list of all excluded foods (Supplementary Excel File 1).

ii) Categorising Foods into Practical LC Food Groups

We used an iterative process to categorise the identified LC foods into practical LC food groups. The top 10 food sources for each nutrient were merged in Excel and automatically sorted according to food type. Two researchers (JT and RF) manually sorted individual foods into groups of similar food types until all foods were grouped. The smallest number of food groups that allowed for variable dietary preferences and food availability was used.

iii) Developing a Sample Meal Plan for a VLC Diet

A 3-day meal plan was developed by a registered dietitian (JT) to be suitable for a VLC diet containing ~30g/day of digestible carbohydrates. The LC food groups were used to construct individual meals, and only foods from the top 10 LC food lists were included in the meal plan, with the exception of some minimally processed added fats (i.e., butter, olive oil and coconut cream) included to help meet energy requirements. The energy target for the meal plan was set at 9,088kJ/day (2,172kcal/day) which reflects minimum energy requirements for most adults to maintain a healthy weight (BMI = 22kg/m2) [18]. The meal plan was designed to meet the upper end of the AMDR for protein (15-25% energy) [18], with remaining energy to be made up of dietary fats. The 3-day meal plan was inputted into FoodWorks (Xyris software, version 10) with foods selected from the Australian Food Composition Database [21]. The nutrient adequacy of the meal plan was assessed to calculate the average amount per day of all essential nutrients analysed. The total daily amounts of dietary energy, digestible carbohydrates, fibres (non-digestible carbohydrates), proteins, and fats in the meal plan were also assessed and reported.

Results

Table 1 presents the top 10 LC food sources of 27 essential micronutrients. Twelve (12/27) nutrients had all 10 LC food sources containing ≥20% NRV of that nutrient per serve for both men and women (Table 1). These nutrients were vitamin D, vitamin B2, vitamin B6, vitamin B7, vitamin B12, copper, iodine, molybdenum, phosphorus, selenium, sodium, and total long-chain omega-3 fats (Table 1). Of the remaining nutrients, nine (9/27) had all 10 LC foods that contained ≥20% NRV for women but not for men, and six (6/27) nutrients had <10 LC food sources for both women and men (Table 1). Vitamin A, potassium, and fluoride had the lowest number of LC food sources containing ≥20% NRV of that nutrient per serve with each having five or less for both men and women (Table 1). Three (3/27) nutrients (vitamin A, vitamin B12 and zinc) had 100% animal food sources, and 16 nutrients (16/27) had >80% animal food sources (Table 1). One nutrient (vitamin C) had 100% plant food sources, and two additional nutrients (vitamin B7 and magnesium) had >80% plant food sources (Table 1). Eighteen LC food groups were identified and are presented in Table 2 with the individual foods included in each group. The LC food group representing the highest number of nutrients was oily fish (20 different nutrients, n=20), followed by seafood (n=16), offal (n=14), white fish (n=11), nuts (n=11), legumes (n=11), seeds (n=11), eggs (n=10), poultry (n=9), noncruciferous vegetables (n=9), pork (n=7), ruminant animals (n=7), game meat (n=7), cruciferous vegetables (n=7), dairy (n=6), fruit (n=5), fungi (n=5), then grains (n=3) (Table 2). The 3-day sample meal plan is presented in Table 3, and its nutrient analysis is presented in Supplementary Table 4. The total energy provided in the meal plan is 9,017kJ/day (2,155kcal/day) with 29g/day of carbohydrates (5% TEI), 128 g/day of proteins (24% TEI), and 163 g/day of fats (67% TEI) (Supplementary Table 4). The meal plan met or exceeded the highest NRVs by ≥90% for 24/27 and 25/27 nutrients, for men and women respectively. The male NRVs that were <90% met were chromium (>50% AI), potassium (83% SDT), and total long-chain omega-3 fats (85% SDT). The female NRVs that were <90% met were chromium (>70% AI) and potassium (83% SDT) (Supplementary Table 4). Nutrient data in the FoodWorks database was incomplete for vitamin B5, vitamin B7, chromium, copper, fluoride, manganese, and molybdenum, so values are likely underestimates.

Table 1: Top 10 low-carbohydrate food sources of essential micronutrients available from the Australian Food Composition Database.

Note: *cooked weight. Abbreviations: AI: Adequate Intake; DHA: Docosahexaenoic Acid; DPA: Docosapentaenoic Acid; EPA: Eicosapentaenoic Acid; kJ: Kilojoules; kcal: kilocalories (kcal=kJ/4.184); med: Medium; tbs: Tablespoons; M: Men; W: Women; mg: Milligrams; NRV: Nutrient Reference Value; RDI: Recommended Dietary Intake; SDT: Suggested Dietary Target; ug: micrograms. .

Table 2: Key nutrients provided in low-carbohydrate food groups.

Note: *cooked weight. Abbreviations: AI: Adequate Intake; DHA: Docosahexaenoic Acid; DPA: Docosapentaenoic Acid; EPA: Eicosapentaenoic Acid; kJ: Kilojoules; kcal: kilocalories (kcal=kJ/4.184); med: Medium; tbs: Tablespoons; M: Men; W: Women; mg: Milligrams; NRV: Nutrient Reference Value; RDI: Recommended Dietary Intake; SDT: Suggested Dietary Target; ug: micrograms. .

Table 3: Sample 3-day meal plan for a very low-carbohydrate diet (total digestible carbohydrates: ≤30g/day).

Discussion

This research successfully identified the top LC food sources and food groups for 27 essential micronutrients using the FSANZ Australian Food Composition Database. This study is the first of its kind to present practical information on meeting the NRVs for essential vitamins, minerals, and fatty acids using a LC dietary approach based on whole foods. The results of this study can be applied in clinical practice by healthcare practitioners to guide the development of LC dietary plans and educational resources for patients. The results can also be applied in future research studies investigating LC diets wherein nutritional adequacy is important (e.g., diet interventions >2 weeks).

It appears that the prioritisation of animal-based protein foods, including meat, fish, seafood, eggs, and dairy, is a critical consideration for meeting micronutrient requirements on a LC diet. Despite using the serve sizes established in the ADGs which recommend larger serves of plant-based proteins (i.e., 150 g) and smaller serves of animal-based proteins (i.e., 65-80 g) [16], over half of the nutrients assessed had ≥80% animal foods as their top LC dietary sources, and most of the LC food groups identified were animal foods. As such, seven (7/9) meals in the 3-day VLC meal plan contained ≥150 g (cooked weight) of animal-based proteins. This poses practical challenges and safety concerns regarding nutritional deficiencies for individuals choosing to limit or exclude animal foods while following a LC diet. In these cases, inclusion of ultra-processed plant-based meat and dairy alternatives that have been fortified with key nutrients (e.g., potassium, calcium, vitamin B12, iron) and/or targeted nutritional supplementation would need to be considered.

With that said, high intakes of animal protein foods may raise concern for disease risk, especially with consideration of the associations between red meat and colorectal cancer [23]. However, flaws identified in this associational data have led healthcare practitioners to question whether the small absolute risk linking red meat to disease (51 people per 100,000 vs. 43 people per 100,000 [23]) is of any real clinical significance. Additionally, the research did not distinguish between the different types of red meat consumed, such that minimally processed red meats (e.g., beef mince, lamb shoulder, pork chop) and naturally preserved meats (i.e., meat that has been transformed through salting, curing, fermentation, etc.) are categorised the same as ultra-processed red meat products (e.g., party pies, sausage rolls, pizza with meat toppings, etc.) [23]. A 2020 analysis (n = 7411) found that Australians in the highest quintile of ultra-processed food consumption had significantly higher odds of having obesity and abdominal obesity compared with those in the lowest quintile of consumption [24], with obesity established as a major risk factor for colorectal cancer [25]. Given that our analysis excluded ultra-processed foods (NOVA 4) [22], the potentially negative health effects of prioritising animal-based protein foods, including red meat, are likely minimized.

Our research suggests that it may be difficult to meet nutritional adequacy for certain nutrients, including vitamin A, fluoride, and potassium, on a LC diet using single food sources. For example, potassium had zero LC foods containing ≥20% NRV per serve. However, the 3-day meal plan still provided 83% of the SDT for potassium (men and women: 4700mg) since a wide variety of whole foods contain potassium and many single food sources add up to a significant amount across a full day. Similarly, the meal plan provided >100% of the NRV for both men and women for vitamin A and fluoride. It is also of important note that, unlike previous research assessing nutritional adequacy of LC diets [19,20], we used the highest available NRVs, including SDTs. This conservative approach made it less attainable for our 3-day meal plan to meet nutrient targets for nutrients with an SDT, given SDTs are typically set much higher than RDIs or AIs [18]. Nevertheless, our 3-day meal plan met or exceeded the SDTs for vitamin A (retinol), vitamin E, vitamin B9 (folate), vitamin C, sodium, and long-chain omega-3 fats (for women).

There are several strengths of this research. The FSANZ Australian Food Composition database analysed is a validated food database that includes a large variety of foods commonly available and consumed in Australia [21]. The systematic approach used to review this database against pre-specified criteria to identify the top LC food sources of each nutrient minimized the risk of researcher bias. In addition, the decision to merge similar foods allowed for increased variation in the types of LC food sources included, accommodating for individual dietary requirements and food preferences. For example, the current list for long-chain omega-3 fats includes liver, seaweed, and mutton in addition to fish and seafood. If we did not merge similar foods, then the list would have only included different species of oily fish.

Limitations of this research are inherent to the nutrient data that was available in the Australian Food Composition Database [21]. Data for vitamin K, choline and linoleic acid was not available. Nevertheless, the USDA food database [26] shows that good sources of vitamin K and choline include whole eggs, dairy, offal, and ruminant animals, which were practical LC food groups identified in the current study. There were also some known nutrient-dense whole foods that were not available in the database, such as salmon roe and lamb’s brain, which would have likely been identified as top sources of one or more essential nutrients in this study. However, it is presumed that these foods are not commonly consumed amongst Australian diets and thus, their lack of inclusion in the database has not largely impacted our results. Finally, The Australian Food Composition database was specific to foods commonly available within Australia and many international foods were not able to be included and assessed. Future research studies assessing nutritional adequacy of LC diets should consider common food items across a variety of countries.

Conclusion

In conclusion, this study successfully identified the top LC food sources and food groups for 27 essential micronutrients using the FSANZ Australian Food Composition Database. Dietitians and other healthcare practitioners can use the results of this research to develop LC dietary plans that meet or exceed the NRVs for vitamins, minerals and fatty acids. The results may also be applied in the design of future experimental studies to guide the formulation of nutritionally adequate LC diet interventions.

Acknowledgements

Funding

None.

Conflicts of Interest

None

Author Contributions

• Jessica Turton: Conceptualization, Methodology, Formal analysis, Data Curation, Writing - Original Draft, Writing - Review & Editing

• Rowena Field: Conceptualization, Methodology, Formal analysis, Data Curation, Writing - Original Draft, Writing - Review & Editing

• Noor Struik: Conceptualization, Methodology, Writing - Review & Editing

• Kieron Rooney: Writing - Review & Editing, Supervision

• Helen Parker: Writing - Review & Editing.

References

1. Turton J, Brinkworth GD, Field R, Parker H, Rooney K (2019) An evidence-based approach to developing low-carbohydrate diets for type 2 diabetes management: A systematic review of interventions and methods. Diabetes, Obesity and Metabolism 21(11): 2513-2525.
2. Goldenberg JZ, Day A, Brinkworth GD, Sato J, Yamada S, et al. (2021) Efficacy and safety of low and very low carbohydrate diets for type 2 diabetes remission: systematic review and meta-analysis of published and unpublished randomized trial data. BMJ 372: m4743.
3. Paoli A, Mancin L, Giacona MC, Bianco A, Caprio M (2020) Effects of a ketogenic diet in overweight women with polycystic ovary syndrome. Journal of Translational Medicine 18(1): 104.
4. Li J, Bai WP, Jiang B, Bai LR, Gu B, et al. (2021) Ketogenic diet in women with polycystic ovary syndrome and liver dysfunction who are obese: A randomized, open-label, parallel-group, controlled pilot trial. Journal of Obstetrics and Gynaecology Research 47(3): 1145-1152.
5. Porchia LM, Hernandez Garcia SC, Gonzalez Mejia ME, López Bayghen E (2020) Diets with lower carbohydrate concentrations improve insulin sensitivity in women with polycystic ovary syndrome: A meta-analysis. European Journal of Obstetrics & Gynecology and Reproductive Biology 248: 110-117.
6. Turton JL, Raab R, Rooney KB (2018) Low-carbohydrate diets for type 1 diabetes mellitus: A systematic review. PLOS ONE 13(3): e0194987.
7. Lennerz BS, Barton A, Bernstein RK, Dikeman RD, Diulus C, et al. (2018) Management of Type 1 Diabetes With a Very Low–Carbohydrate Diet. Pediatrics 141(6): e20173349.
8. Hyde PN, Sapper TN, Crabtree CD, LaFountain RA, Bowling ML, et al. (2019) Dietary carbohydrate restriction improves metabolic syndrome independent of weight loss. JCI insight 4(12): e128308.
9. Gershuni VM, Yan SL, Medici V (2018) Nutritional Ketosis for Weight Management and Reversal of Metabolic Syndrome. Curr Nutr Rep 7(3): 97-106.
10. Sackner Bernstein J, Kanter D, Kaul S (2015) Dietary Intervention for Overweight and Obese Adults: Comparison of Low-Carbohydrate and Low-Fat Diets. A Meta-Analysis. PloS one 10(10): e0139817.
11. Kishk NA, Yousof HZ, Ebraheim AM, Elkholy TAFA, Soliman SH, et al. (2021) The effect of ketogenic diet escalation in adolescents and adults with drug-resistant epilepsy: a prospective study. Nutritional Neuroscience 20: 1-10.
12. Christensen MG, Damsgaard J, Fink Jensen A (2021) Use of ketogenic diets in the treatment of central nervous system diseases: a systematic review. Nordic Journal of Psychiatry 75(1): 1-8.
13. Neves GS, Lunardi MS, Lin K, Rieger DK, Ribeiro LC, et al. (2020) Ketogenic diet, seizure control, and cardiometabolic risk in adult patients with pharmacoresistant epilepsy: a review. Nutrition Reviews 79(8): 931-944.
14. Churuangsuk C, Griffiths D, Lean MEJ, Combet E (2019) Impacts of carbohydrate-restricted diets on micronutrient intakes and status: A systematic review. Obesity reviews: an official journal of the International Association for the Study of Obesity 20(8): 1132-1147.
15. (2011) National Health and Medical Research Council. A Modelling System to Inform the Revision of the Australian Guide to Healthy Eating. Commonwealth of Australia.
16. (2013) National Health and Medical Research Council. Australian Dietary Guidelines. Commonwealth of Australia.
17. (2021) National Health and Medical Research Council. Australian Guide to Healthy Eating.
18. (2006) National Health and Medical Research Council. Nutrient reference values for Australia and New Zealand: including recommended dietary intakes Canberra, A.C.T: National Health and Medical Research Council.
19. Zinn C, Rush A, Johnson R (2018) Assessing the nutrient intake of a low-carbohydrate, high-fat (LCHF) diet: a hypothetical case study design. BMJ Open 8(2): e018846.
20. Jebeile H, Grunseit AM, Thomas M, Kelly T, Garnett SP, et al. (2020) Low-carbohydrate interventions for adolescent obesity: Nutritional adequacy and guidance for clinical practice. Clinical Obesity 10(4): e12370.
21. (2019) Food Standards Australia & New Zealand. Australian Food Composition Database.
22. Monteiro CA, Cannon G, Lawrence M, Costa Louzada ML, Pereira Machado P (2019) Ultra-processed foods, diet quality, and health using the NOVA classification system. Rome: Food and Agriculture Organization of the United Nations.
23. Bouvard V, Loomis D, Guyton KZ, Grosse Y, Ghissassi FE, et al. (2015) Carcinogenicity of consumption of red and processed meat. The Lancet Oncology 16(16): 1599-1600.
24. Machado PP, Steele EM, Levy RB, da Costa Louzada ML, Rangan A, et al. (2020) Ultra-processed food consumption and obesity in the Australian adult population. Nutrition & Diabetes 10(1): 39.
25. Ma Y, Yang Y, Wang F, Zhang P, Shi C, et al. (2013) Obesity and risk of colorectal cancer: a systematic review of prospective studies. PloS one 8(1): e53916.
26. (2019) United States Department of Agriculture. USDA Food Composition Databases.

ABSTRACT

Background: There is a lack of evidence-based resources available to guide the formulation of nutritionally adequate low-carbohydrate (LC) diets.

Methods: We searched the FSANZ Australian Food Composition Database to identify the top 10 LC food sources for each of the 27 essential micronutrients (vitamins, minerals, and fatty acids). To be included, foods had to contain ≤5.0g total digestible carbohydrates per serve (serving), contain at least 20% of the highest available Nutrient Reference Value (NRV) per serve, and be classified as a minimally processed food. We then categorized the foods into practical food groups and developed a sample meal plan suitable for a very LC diet (<30g/day) using these resources.

Results: Twelve (12/27) nutrients had 10 LC foods containing ≥20% NRV of that nutrient per serve for both men and women. Nine (9/27) nutrients had 10 LC foods containing ≥20% NRV for women but not for men, and six (6/27) nutrients had <10LC food sources for both women and men. Vitamin A, vitamin B12 and zinc had entirely animal food sources, and 16/27 nutrients had >80% animal food sources. Vitamin C had entirely plant food sources, and two additional nutrients (vitamin B7 and magnesium) had more than 80% plant food sources. The top 10 LC foods across all 27 nutrients were categorised into 18LC food groups. The group representing the highest number of nutrients was oily fish followed by seafood, offal, white fish, nuts, legumes, seeds, eggs, poultry, non-cruciferous vegetables, pork, ruminant animals, game meat, cruciferous vegetables, dairy, fruit, fungi, then grains. The sample meal plan met or exceeded the highest NRVs by ≥90% for 24/27 and 25/27 nutrients, for men and women respectively.

Conclusion: This study presents practical information to support healthcare practitioners and future researchers in formulating nutritionally adequate LC dietary plans.

Keywords: Diet Therapy; Diet; Carbohydrate- Restricted; Micronutrients; Nutrients; Nutritionists

Abbreviations: LC: Low-Carbohydrate; VLC: Very Low-Carbohydrate; ADGs: Australian Dietary Guidelines; AGHE: Australian Guide to Healthy Eating; TEI: Total Energy Intake; NRVs: Nutrient Reference Values; RDI: Recommended Dietary Intakes; AI: Adequate Intakes; FSANZ: Food Standards Australia New Zealand; SDT: Suggested Dietary Target