Chemical food composition: implications for atherosclerosis prevention

Abstracts

OBJETIVO: Comparar a composição de ácidos graxos e colesterol de alimentos adquiridos no mercado brasileiro, com tabelas de referência mais usadas no Brasil. MÉTODOS: Foi analisada a composição de ácidos graxos e colesterol de 41 alimentos mais usados na alimentação em nosso meio e modos de preparo, utilizando-se metodologias específicas e comparadas com as informações das tabelas adotadas pela Unicamp e UNIFESP. RESULTADOS: O conteúdo de colesterol encontrado no queijo parmesão foi de 100,7 mg/100 g; já na tabela da UNIFESP foi de 68 mg/100 g (p < 0,05), ou seja, 48% a mais. Para o queijo tipo prato, a tabela deste estudo apresentou medida 31% menor de colesterol (94 mg/100 g versus 123 mg/100 g, p < 0,05). Para o leite integral, a variação é de 52% com relação ao colesterol, enquanto a gordura saturada variou de 1,4 g/100 g, na tabela da Unicamp, a 2,130 g/100 g, na tabela do estudo (p < 0,05). Para alguns alimentos, não foram encontradas diferenças estatisticamente significativas entre as tabelas avaliadas. Entretanto, quando foi prescrita uma dieta de 1.800 calorias, as discrepâncias entre as tabelas e a falta de informações resultaram em diferenças clinicamente relevantes nas recomendações dietéticas. CONCLUSÃO: Existem importantes diferenças no teor das gorduras nos alimentos avaliados quando comparados com as tabelas mais usadas em nosso meio, o que pode comprometer as recomendações para a prevenção da aterosclerose. Uma das explicações seria o fato de a tabela da UNIFESP ser de origem americana.

Dieta; dislipidemias; fatores de risco; colesterol; ácidos graxos

ORIGINAL ARTICLE

Chemical food composition: implications for atherosclerosis prevention

Carlos Scherr^I; Jorge Pinto Ribeiro^II

^IPh.D. in Cardiovascular Sciences; Scientific Director, Instituto do Coração e do Diabetes, Rio de Janeiro, RJ, Brazil

^IIFaculty member; Chief, Graphic Methods, Cardiology Service, Hospital Central da Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brazil

^{Correspondence to}

SUMMARY

OBJECTIVE: To compare the fatty acid and cholesterol content in food acquired in Brazil with the composition found in the most frequently used reference tables in the country.

METHODS: The fatty acid and cholesterol content in 41 food items frequently used in our country and the various directions to prepare them were reviewed by using specific methodology and the information was compared to the tables adopted by Unicamp and UNIFESP.

RESULTS: According to Unicamp table, the cholesterol content found in parmesan cheese was 100.7 mg/100 g, while it was 68 mg/100 g in UNIFESP table, that is, a 48% (p < 0.05), higher content in the former. This study table found a cholesterol content 31% lower (94 mg/100 g vs. 123 mg/100 g, p < 0.05) for yellow cheese. For whole milk, we found a 52% difference regarding cholesterol content, while the difference for saturated fat ranged from 1.4 g/100 g in Unicamp table to 2.130 g/100 g in our study table (p < 0.05). For some food items, no statistically significant differences were found among the tables. However, when a 1,800-calorie diet was prescribed, the discrepancies among the tables and lack of information resulted in clinically relevant differences in dietary recommendations.

CONCLUSION: There are important differences in food fat content between the fatty acid and cholesterol content formally analyzed and the content shown on commonly used tables, and this can compromise our recommendations on preventing atherosclerosis. One possible explanation for the differences would be the fact that the UNIFESP table is American in origin.

Keywords: Diet; dyslipidemias; risk factors; cholesterol; fatty acids.

Introduction

The World Health Organization projections show a greatly increased mortality from cardiovascular diseases in developing countries, Brazil included, and a relenting behavior in industrialized countries¹. Risk factors, such as cigarette smoking, low physical activity and unhealthy food are directly linked to these changes. Several authors have shown the advantages of controlling risk factors in studies with thousands of individuals experiencing coronary artery disease or not. In 1990, Ornish et al.² showed the results of vegetarian diet, smoking quitting, emotional stress control techniques, and moderate physical activity over one year in a small group with 28 patients having coronary artery disease. A coronary angiography performed before the study and repeated one year later revealed the stenosis grade was partially decreased in 82% of the lesions in the group undergoing intervention and mainly for the patients with more severe lesions, while progression was seen in the control group. Therefore, the study by Ornish et al.² demonstrated a dietary intervention with no medication can result in important effects on the coronary artery disease. The Lyon Diet Heart Study also showed reduced mortality and myocardial infarction in those who adhered to Mediterranean style diet, compared to the usual diet, with no significant change in blood pressure or even in blood lipids for these patients, what makes us believe the very type of food played an independent role in the outcome³.

To achieve both a food inquiry and a diet prescription, a food content table lookup is required. In Brazil, one of the most used references is the Brazilian Food Composition Table, designed from food industry information based on strict analysis criteria through completing a special form created by the Food and Experimental Nutrition Department, Pharmaceutical Sciences, Universidade de São Paulo (USP)⁴. However, this table is limited, as it only provides data on total lipids and cholesterol concerning fatty acids and cholesterol. It does not inform either about different ways of preparing the food in some types of meat. Another common source, the table used by Escola Paulista de Medicina, Universidade Federal de São Paulo (UNIFESP)⁵, was indeed adapted from a table developed by the United States Department of Agriculture, Agricultural Research Service, 2001, therefore being American. It has information concerning total lipids, cholesterol, saturated, mono- and polyunsaturated fats, but the information about preparation methods is not clear⁵.

The Brazilian most complete food composition reference is likely the Brazilian Food Composition Table, designed by the Food Study and Research Nucleus from Universidade Estadual de Campinas (NEPA/Unicamp)⁶. The content of total lipids, cholesterol, saturated, poly- and monounsaturated fats is found in it, as well as all the fatty acid chain. However, fish is only presented as raw fish; beef strip loin can be found only grilled or raw, chicken breast is only presented raw; skim milk can only be found powdered and there is no semi. Finally, data about cheese, butter, margarines, sausage meat groups, edible oils and others are not available⁶.

In addition, the food chemical composition can be different according to the reference table⁷. Ribeiro et al. ⁷, in 2003, had already shown differences in lipids when they compared the values provided by the tables with those obtained in laboratory. Recently we carried out an extensive evaluation on the chemical composition of a large number of foods used in Brazil to build a food composition table which could be clinically applied in our country⁸. In this current report, we analyze the fatty acid and cholesterol chemical composition in some foods that can be formally bought in Brazil and compare the contents found with those in the most commonly used tables in our country.

Methods

In this cross-sectional study, foods that had comparable matches in other tables, such as tilsit and yellow cheese, farm eggs, low-fat chicken and turkey sausage, whole, semi and skim milk were analyzed.

Food chemical composition analysis - All the food analyses listed below were carried out in the laboratories of Instituto de Tecnologia de Alimentos (Food Technology Institute - ITAL), Agriculture and Supply Secretary, São Paulo State Government and funded by Instituto Nacional de Metrologia, Normalização e Qualidade Industrial (National Metrology, Normalization and Industrial Quality Institute - INMETRO), subordinated to Brazilian Development, Industry and Foreign Trade Ministry. Specific methodology was used to assess the food composition regarding total lipids^9,10, cholesterol^11,12, fatty acid composition¹³, sodium chloride content measurement¹⁴ and moisture^15,16.

All the analyses were conducted using calibrated glassware and equipment. Their uncertainties follow acceptance criteria by ITAL quality system (NBR-ISO 9001) and they were not taken into account for the result calculation, but they are available for consultation. The analyses were carried out both on a wet basis and on a dry basis, but only the measurements on a wet basis were considered, since they represent the manner the food is usually consumed. According to the governmental regulation 27 of January 13, 1998 by the Health Surveillance Secretary of the Health Ministry, trans fatty acids should be computed into the saturated fat calculation.

Sample preparation

Two to five different brands of semi, skim, whole, and omega 3-containing milks were acquired to be analyzed. Six brands of Minas e yellow cheeses, as well as five brands of butter and margarine were analyzed. Eggs whose package indicated factory-farmed, free-range or low-fat types were also evaluated. Each sample consisted of six eggs boiled for 10 minutes and then homogenized and weighed, being the egg white finally separated.

The statistical analysis of the Scherr Table to validate the fat content in foods acquired and produced in Brazil was conducted by the Mann-Whitney test (non-parametric test) to compare acid, lipid, and cholesterol measurements among different foods analyzed. The significance criterion adopted was the 5% level.

Results

Forty-one different food products were analyzed and a total of 733 comparisons were made among them. Only those present in at least one of the reference tables are shown.

Differences between this study table and the UNIFESP table were observed for parmesan cheese fat content, ranging from 100.7 mg/100 g to 68 mg/100 g (p < 0.05), that is, 48% higher difference. In the case of saturated fat, the difference was lower - from 8.00 g/100 g to 16.77 g/100 g (p < 0.05), respectively. Making the same comparison for yellow cheese, this study table showed a 31% lower cholesterol measurement (94 mg/100 g vs 123 mg/100 g, p < 0.05). As demonstrated in Table 1, for whole milk the variation is of 52% regarding cholesterol, while saturated fat changed from 1.4 g/100 g in Unicamp table to 2.130 g/100 g in our study table (p < 0.05). Regarding 1% fat skim milk, the cholesterol content variation was 71.6% (2.91 mg/100 g in the study table and 5 mg/100 g in UNIFESP table, p < 0.05).

For some of the foods, no statistically significant differences were found among the analyzed tables. As an example, Table 2 shows this study table compared to UNIFESP table for chicken sausage, demonstrating similar composition. However, when we proposed a 1,800-calorie menu based on American Heart Association recommendations, the great lack of information and the discrepancies in cholesterol and fatty acid contents, depending on the table applied, became evident. For a similar menu, total lipid contents of 35.87, 29.94, and 208.3 and a cholesterol content of 299.58, 14.0 and 247.0 were found, depending on the query being performed on our study table, TACO study table or UNIFESP table, respectively.

Discussion

Whether from missing information or from no adaptation to real products delivered to Brazilian population, there is a considerable gap to be filled in the cholesterol and fatty acid chemical composition issue in the tables available to date. Acknowledging this fact, it is possible to deliver more reliable, attractive, and non-punishing menu options, leading to a higher adherence to a heart healthier food intake, considering the low adherence to any kind of diet within three months¹⁷.

The food chemical composition tables most frequently used in our country do not provide information on cholesterol and fatty acids for most foods in the way they are consumed. Although TACO⁶, used by Unicamp, is the most complete, listing a number of meats and their different cuts, it does not provide data related to the preparation method and regards only the raw presentation. On the other hand, UNIFESP table⁵ is American in origin and although its data are more comprehensive, they do not mirror the national reality, in which the cattle has different breeds and is differently fed. Finally, USP⁴ uses data gathered from a well-structured questionnaire completed by the food producers, but it does not include the information on the preparation method either. Our data regards food produced and consumed in our own country, with the samples being collected in the formal market where housewives and dealers buy their products. As many authors have already demonstrated^18-20, there are many different influences on meat cholesterol and chemical composition from cooking methods. Rosa et al.²¹ came to the same conclusion by comparing the effects of cooking in water, oil, grill, standard oven or microwave oven regarding chicken breast and leg fats, finding differences based on the preparation method. These studies show oil-free cooking brings about lipid loss, whereas fried foods lead to oil absorption, with a cut-dependent absorption²⁰.

Many studies have proved the food influence on blood lipids and atherosclerosis course^22,23 and other studies show it is difficult to make the population improve their eating habits¹⁷. If, instead of mentioning what must not be eaten, we show simple solutions that might make adherence easier for the general population and particularly for the patients and are based on clear and reliable information, dietary interventions will likely be more effective. In practice, this study targets to increase the awareness of further information necessity about the cholesterol and fatty acids content under two aspects: the first regards the items used by Brazilian population; the second regards the chemical composition after the food is handled and prepared to be consumed at the table.

This study is only focused on food chemical composition regarding cholesterol and the fatty acid chain and does not reflect all the components in Brazilian food chain or all the usual preparation methods. When it is compared to the available information on the most used tables in our country, the lack of a great deal of information is found in the latter ones or, when available information is compared to the data achieved in this study, great differences are found, mostly when our data is compared to foreign tables.

Conclusion

The current analysis demonstrates great differences between the chemical composition of some foods and the contents shown in national reference tables, suggesting nutritional intervention has possibly been impaired in our country.

References

Submitted on: 08/03/2010

Approved on: 12/27/2010

Conflicts of interest: None.

Funding: This study was funded by the Instituto de Metrologia e Estatística - Ministério do Desenvolvimento, Indústria e Comércio Exterior [Metrology and Statistical Institute - Development, Industry and Foreign Trade Ministry].

From Universidade Gama Filho, Instituto do Coração e Diabetes, Rio de Janeiro, RJ; Cardiology Service, Hospital de Clínicas de Porto Alegre and Internal Medicine Department, Medical School, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brazil.

• 1. Reddy KS. Cardiovascular disease in non-Western countries. N Engl J Med. 2004;350:2438-40.
• 2. Ornish D, Brown SE, Sherwitz LW, Billings JH, Armstrong WT, Ports TA et al Can lifestyle changes reverse coronary heart disease? The Lifestyle Heart Trial. Lancet 1990;336:129-33.
• 3. De Lorgeril M, Salen P, Martin JL, Monjaud I, Delaye J, Mamelle N. Mediterranean diet, traditional risk factors, and the rate of cardiovascular complications after myocardial infarction: final report of the Lyon Diet Heart Study. Circulation 1999;99:779-85.
• ⁴
  Universidade de São Paulo. Faculdade de Ciências Farmacêuticas. Departamento de Alimentos e Nutrição Experimental/BRASILFOODS (1998). Tabela brasileira de composição de alimentos-USP. Versão 5.0. [citado 17 out 2010]. Disponível em: http://www.fcf.usp.br/tabela
• 5. Universidade Federal de São Paulo. Departamento de Informática em Saúde. Tabela de composição química dos alimentos (2001). [citado 17 put 2010]. Disponível em: http://www.unifesp.br/dis/servicos/nutri
• ⁶
  Universidade Estadual de Campinas. Núcleo de Estudos e Pesquisas em Alimentação. TACO. Tabela brasileira de composição química de alimentos-Versão 2. [citado 18 out 2010]. Disponível em: http://www.unicamp.br/nepa/taco
• 7. Ribeiro P, de Morais TB, Colugnati FA, Sigulem DM. Tabelas de composição química de alimentos: análise comparativa com resultados laboratoriais. Rev Saúde Pública 2003;13:39-43.
• 8. Scherr C, Ribeiro JP. Redução do risco cardiovascular. Nova tabela de composição de colesterol e ácidos graxos em alimentos. Rio de Janeiro: Guanabara Koogan; 2009.
• 9. Folch J, Lees M, Stanley GHS. A simple method for the isolation and purification of total lipíds from animal tissues. J Bioll Chem. 1957;226:497-509.
• 10. Gerber A, Van Gullik D. Normas analíticas do Instituto Adolfo Lutz. São Paulo: Instituto Adolfo Lutz; 1985. p. 270.
• 11. Bragagnolo N, Rodriguez-Amaya DB. Avaliação comparativa de três métodos para determinação de colesterol em gema de ovo. Arq Biol Tecnol. 1993;36:237-51.
• 12. Schmarr H, Gross HB, Shibamoto Analysis of polar cholesterol oxidation products: evaluation of a new method involving transesterification, solid phase extraction, and gas chromatography. J Agric Food Chem. 1996;44:512-17.
• 13. Firestone D. Official methods and recommended practices of the American Oil Chemists Society. Champaign: American Oil Chemists Society, 2008. p.11-96.
• 14. Helrich K. Official methods of analysis of the Association of Official Analytical Chemists. Arlington: Association of Official Analytical Chemists, 1990. p.842.
• 15. Silva PHF, Pereira DBC, Oliveira LL, Costa Jr. LCG. Físico-química do leite e derivados: métodos analíticos. Juiz de Fora: Editora, Oficina de Impressão Gráfica, 1997. p.28-9.
• 16. Cunniff P. Official methods of analysis of the Association of Official Analytical Chemists. Arlington: Association of Official Analytical Chemists, 1998.
• 17. Dansinger ML, Gleason JA, GriffithJL, Selker HP, Schaefer EJ. Comparison of the Atkins, Ornish, Weight Watchers, and Zone diets for weight loss and heart disease risk reduction: a randomized trial. JAMA 2005;293:43-53.
• 18. Garcia-Arias MT, Pontes EA, Garcia Linhares MC, Fernandez MCG, Sanchez-Muniz FJ. Cooking-freezing-reheating (CFR) of sardine (Sardina pilchardus) fillets: effect of different cooking and reheating procedures on the proximate and fatty acid compositions. Food Chem. 2003;83:349-56.
• 19. Potter NN, Hotchkiss JH. Ciência de los alimentos. Zaragoza: Acribia, 1995. p. 667.
• 20. Gokolu N, Yerlikaya P, Cengiz E. Effects of cooking methods on the proximate composition and mineral contents of rainbow trout (Oncorhynchus mykiss). London: Food Chemistry, 2003.
• 21. Rosa FC, Bressan MC, Bertechini AG. Efeito de métodos de cocção sobre a composição química e colesterol em peito e coxa de frangos de corte. Ciênc Agrotec. 2006;30:707-14.
• 22. Denke MA. Cholesterol-lowering diets. A review of the evidence. Arch Intern Med. 1995;155:1684-5.
• 23. Truswell AS. Review of dietary intervention studies: effect on coronary events and on total mortality. Aust N Z J Med. 1994;24:98-106

Publication Dates

• Publication in this collection
  30 May 2011
• Date of issue
  Apr 2011

History

• Accepted
  27 Dec 2010
• Received
  03 Aug 2010