ORIGINAL / Nutrición enteral

 

ILSI Task Force on enteral nutrition; estimated composition and costs of blenderized diets

Grupo de estudios ILSI sobre nutrición enteral; composición y costos estimados de dietas artesanales

 

 

Roseli Borghi¹, Thalita Dutra Araujo², Roberta Ianni Airoldi Vieira³, Telma Theodoro de Souza⁴ and Dan Linetzky Waitzberg⁵

¹Dietitian, Scientific Coordinator of Nestle Health Science
²Nutrition Intern of Nestlé Health Science
³Dietitian, Specialist in Exercise Physiology, Scientific Advisor of Support Advanced Medical Nutrition (A Danone Group Company)
⁴Dietitian, Specialist in Exercise Physiology, Stakeholder Manager of Support Advanced Medical Nutrition (A Danone Group Company)
⁵MD, Associate Professor of the Department of Gastroenterology. Sao Paulo University Medical School. Director of Ganep Human Nutrition. Sao Paulo. SP. Brasil

This work was supported by the Task Force on Clinical Nutrition of the International Life Sciences Institute Brasil (ILSI Brasil). Industry members of the Committee include Abbott Laboratôrios do Brasil Ltda., Danone Ltda., DuPont do Brasil, Mead Johnson Nutrition, Nestlé Brasil Ltda., Support Produtos Nutricionais Ltda., and Unilever Brasil Ltda. (For further information about the Task Force or ILSI Brasil, call (+55 11) 3035-5585 or email ilsibr@ilsi.org.br). The opinions expressed herein are those of the authors and do not necessarily represent the views of ILSI Brasil.

Correspondence

 

 

---

ABSTRACT

Blenderized tube diets (BTD) are used in some parts of Brazil and few studies have analyzed their features in comparison with industrialized preparations. Among 14 randomly collected BTD recipes 9 were poorly described or failed to standardize foodstuffs and portions and, consequently, nutrient and energy composition was difficult to define. Only five BTD allowed theoretical estimation of their nutritional properties. Macronutrient content was highly variable, often conflicting with accepted daily recommendations. According to the literature there are further disadvantages with BTD use including diet high risk of contamination, physical and chemical instability, and high osmolarity and viscosity. Nominal cost of BTD was comparatively low in relation to industrialized formulas; however we did not compute labor and indirect expenses, probably rendering final value more expensive than with the industrialized alternative. It is likely that within such circumstances, hospital and home care malnutrition will not be adequately dealt with and related complications may occur. It is concluded that the continued use of blenderized tube feeding diets requires careful assessment, prioritizing correction of potencial nutritional deficits by means of safe, balanced, chemically complete and effective nutritional prescriptions.

Key words: Enteral nutrition. Blenderized formulas. Hospital-prepared tube feeding. Malnutrition. Costs.

---

RESUMEN

Dietas enterales artesanales (DEA) son utilizadas en algunas partes de Brasil y pocos estudios analizaron sus características en comparación con las preparaciones industrializadas. De un total de 14 DEA colectadas 9 estaban mal definidas o no establecían claramente los géneros alimenticios y porciones empleados, consecuentemente fue difícil calcular su composición en nutrientes y energía. Solo cinco DEA permitieron una estimativa teórica de sus propiedades nutricionales. El contenido de los macronutrientes era muy variable, y con frecuencia divergente de las recomendaciones diarias aceptas. De acuerdo con la literatura hay otros inconvenientes en el uso de DEA, o sea elevado riesgo de contaminación, inestabilidad físico-química, y alta osmolaridad y viscosidad de las DEA. El costo nominal de las dietas fue comparativamente bajo en relación a las formulaciones industrializadas, pero los gastos laborales y otros costos indirectos no fueron computados por nosotros, probablemente conduciendo a un valor total final superior al de la alternativa industrializada. Es esperado que en tales circunstancias la desnutrición no quedara bien tratada y complicaciones relacionadas podrán aparecer. La conclusión es que el empleo continuado de dietas artesanales requiere una evaluación criteriosa, con prioridad para la corrección de los déficits nutricionales utilizando prescripciones seguras, balanceadas quimicamente, completas y efectivas.

Palabras clave: Nutrición enteral. Fórmulas artesanales. Dieta hospitalaria manipulada. Desnutrición. Costos.

---

 

Introduction

Malnutrition is a well known risk for patients treated in hospitals, long-term facilities as well as home care¹. Malnutrition in turn is associated to higher rate of complications, general morbidity and mortality, and prolonged hospitalization when compared to the well-nourished state². Nutritional therapy emerges in this context as an important tool to antagonize hospital malnutrition.

Enteral nutritional therapy (ENT) is widely practiced whenever the patient is unable to use oral feeding, however still enjoys the full or partial ability of absorbing nutrients from the gastrointestinal tract. ENT allows the administration of adequate nutrients thus contributing to alleviate malnutrition with its accompanying infectious complications^3-7. For more than two decades full resources for enteral nutrition practice have been available in Brazil including ready-to-use diets with defined composition, flexible and biocompatible enteral tubes, and custom-designed infusion pumps, thus permitting safe and reliable enteral nutrition therapy with quality control, both in the nosocomial and the home environment^8-9.

Nonetheless, blenderized or homebrew tube diets (BTD) are still popular, especially for home use after hospital discharge as well as during home care by specialized health providers¹⁰. Inside hospitals their use has been decreasing however they are not uncommon in long-term facilities¹¹.

Many defend that BTD are more physiological and cost less than corresponding industrialized enteral diets (IED)^12,13. Yet there is evidence that, in BTD, contamination may be higher, along with physical and chemical instability as well as more cumbersome handling and administration. Microbial colonization may be precipitated during multiple phases encompassing contaminated ingredients and equipment, faulty hygiene of hands and compounding areas, and storage and distribution routines that favor germ proliferation¹⁴. Ideal calorie and protein density is difficult to achieve, whereas macro and micronutrient profile rarely fulfills all requirements of the patients^12-16.

By the same token, in BTD there is little or no control of osmolarity and viscosity, contributing to their inefficiency and complications. When offered in diluted form such diets fail to supply sufficient protein and energy, and if given in a more concentrated way digestive intolerance or tube clogging might occur.

Given the importance of this subject it was selected, for a practical and theoretical analysis, by the Task Force on Clinical Nutrition of the International Life Sciences Institute in Brasil (ILSI Brasil). It is the purpose of such institution to discuss solutions for technical and scientific problems relevant to the population as a whole, integrating industries, government, universities and research institutes, within the context of the rights of each patient to receive optimal nutritional care at all times, and the responsibility of the society and specialized health organizations to provide high quality nutrients, in order to foster health and correct malnutrition.

The objective was a practical and theoretical analysis of the nutritional quality and nominal cost of BDT as adopted in Brazil, in the light of available IED alternatives and international guidelines. End points were macronutrients, selected micronutrients, and adequacy regarding daily requirements of the adult population (dietary reference intakes/DRI and Brazilian recommended daily intake/IDR). Cost assessment was incomplete addressing only price of foodstuffs, leaving out labor and other expenditures.

 

Methods

Recipes of blenderized enteral feeding were randomly collected in 14 hospitals, however only five were utilized, on account of more precise information about foodstuff amounts (weight in g and volume in ml) along with described manipulation routine, indispensable for dietary assessment. The BTD originated from different Brazilian regions, both in the northern and southern parts of the country, but the authors processed their analysis blinded to the BTD provenience.

Energy value was registered in Microsoft Excel spreadsheets and calculated for each ingredient. In case of undisclosed items such as legume soup, Brazilian internet sources were searched for representative composition (http://tudogostoso.uol.com.br/categorias/sopas). Commercial foods were computed according to portion size and manufacturer informations, specifically Mucilon^® Nestlé, Nidex^® Nestlé, Nutren Active^® Nestlé and Sustagen^®, Mead Johnson. For macro- and micro-nutrients the software AVANUTRI PC was employed (Avanutri & Nutrição Serviços e Informática Ltda Me Três Rios/RJ). Results were complemented and confirmed with the help of two Brazilian food composition tables, namely TACO and Phillipi.

Adequacy of macronutrient distribution was classified according to the guidelines of the Brazilian Ministry of Health: carbohydrates: 55-75% of total energy intake (TEI), of which 45-65% complex carbohydrates and fibers, and <10% of free or simple sugars; lipids: 15-30% of TEI; protein: 10-15% of TEI⁷. Protein and vitamin allowances followed the Brazilian IDR as well as the USA DRI.

Foodstuff costs were searched in commercial providers considering the lowest consumer price: www.sonda.com.br; www.farmadelivery.com.br; www.farmaciamix.com.br. Industrialized diets had their monetary value similarly transcribed from internet consumer sites (www.nutriservice.com.br and www.sabordeviver.com.br), focusing the least expensive option.

 

Results

The blenderized tube feeding diets mixed fresh ingredients with selected industrialized complements. All five recipes contained beef or chicken along with legumes. In three cases cooked beans, green leaves and eggs were added, including whole cooked eggs, cooked egg white, or raw egg yolk. Cow milk and in one case soybean milk were also present. Lipids were supplied by means of soybean oil or corn oil, and in one example by vegetable oil plus margarine.

Sugar was identified in one composition, and industrialized maltodextrine (Nidex^®, Nestlé) was part of another one. Additional carbohydrates encompassed oat meal, corn meal or starch, chocolate powder drink, baby cereal (Mucilon^®, Nestlé), and complex nutritional supplements Sustagen^® Mead Johnson and Nutren Active^® Nestlé. Nutritional features of the mixtures can be seen in table I.

Portioning of the diet was not always clear. Two main meals of 300 ml (lunch and dinner) were anticipated along with four smaller meals, to a total of 1400 ml/day (table II).

Aiming to compare the costs with industrialized products, two standard powder preparations were selected. A daily prescription of 1,800 ml (one can) with a calorie density of 1 kcal/ml was initially calculated. The price range for such ready-to-use alternative was 17.55-45.74 Brazilian real (BRL), or approximately 6.5-17 €, with a mean of 31.65 BRL (11.7 €). This value was subsequently corrected to a volume of 1,400 ml (Table III).

 

Direct comparison with industrialized enteral formulas revealed both lower and higher prices for blenderized mixtures, thus it can be accepted that costs of ingredients are roughly comparable. It is important to emphasize that these are incomplete costs overlooking labor, storage, waste disposal, and requirements for water, electricity and other utilities. In this sense, further studies addressing total costs are crucial to envision the full financial impact of the two methods.

Carbohydrate administration was heterogeneous in this series and only one formulation was close to the minimum recommendations. One hospital only achieved adequate protein allowance, and three of the recipes exceeded lipid intake. These findings, taken together, highlight the nutritional inconsistencies in blenderized diets utilized inside and outside the hospital environment (Table IV).

 

Discussion

A variety of dietary compositions emerged from the current analysis. It is worth mentioning that nutritional value of foodstuffs depends on geographical origin, ripeness, season of the year, and methods employed during processing, storage and cooking¹⁸.

Full chemical analysis of blenderized diets could be a useful procedure to establish, even though aproximatelly, the real composition of the BTD feeding. Given the possible external influences already alluded to, discrepancies between lots would still not be ruled out. Nevertheless this would represent an important step, providing health care professionals with fundamental information about the nutritional value of the prescribed diet. Of course such a pitfall doesn t exist with industrialized preparations, which are stable and well defined, thus ensuring that each prescribed nutrient is actually delivered to the patient.

Other authors similarly resent the variability of nutritional composition of blenderized feedings¹². Mokhalalati et al. investigated three homebrew diets in Saudi Arabian hospitals. Not only differences from one institution to another were detected, but also they happened within the same hospital. Fluctuations in the range of 16-50% occurred in the same hospital. Moreover, inadequacies were present regarding 12 nutrients, when compared to the theoretical recipe¹². Another problem concerned energy density, which was variable depending on the availability of foodstuffs. Calculated values did not meet the nutritional requirements of the population¹.

Also in the experience of Henriques et al., low energy density of homebrew diets was a concern¹⁹. The objective of their study was the preparation of diets with adequate osmolarity. Indeed values of 250-400 mOsm/kg were achieved, however calorie density often lagged behind, with a range of 0.60-1.08 kcal/ml corresponding to low to normal energy density according to the ESPEN directives²⁰.

It becomes clear that production of hypercaloric BTD incurs the risk of osmolarity exceeding 400 mOsm/kg and elevated viscosity. Such hypothesis is compatible with the findings of Mokhalalati et al.¹² of viscosity (2,000-4,000 centipoise) and osmolarity (607 ± 204 mOsm/Kg). Industrialized diets typically exhibit values ofjust 10 centipoise, highlighting the virtual impossibility of administering such high viscosity fluids.

One should admit that blenderized tube feeding diets are flexible, permitting free selection of ingredients to reach adequate nutritional value; however this advantage is offset by excessive manipulation during the multiple compounding phases. The consequence is higher microorganisms contamination and also more expenses, as these are labor-intensive steps. Microbiological details were out of the scope of our study and also financial calculations were limited to foodstuff prices only. All this facts notwithstanding, these are relevant points that deserve further studies.

The raw components of the homebrew diet are a natural source of contamination, and aseptic environments, materials and handling procedures during diet compounding cannot always be taken for granted¹⁵. Mauricio et al. noticed 100% contamination by fecal coliforms in BTD from two hospitals, besides molds and fungi in unacceptable concentrations. Both raw materials and excessive manipulation were incriminated, in agreement with other investigations^16,21,22.

Again Freedland et al.²⁴ registered 30-90% contamination in an open system of blenderized diets administered by continuous drip, which was associated with poor aseptic technique, inadequate cleaning and disinfection of compounding equipment, and contaminated ingredients added to the diet. Manipulation and dilution procedures increased the risk of contamination, coinciding with other publications who focus blenderized diet contamination as a risk factor for nosocomial infections^12,25,26.

In addition to susceptibility to bacterial proliferation, blenderized tube feeding diets may cost more. In the current protocol only purchased components were taken in account however, many more items are relevant. Silkroski et al. conducted a survey on costs of tube feeding in academic hospitals in the USA²⁷. They were particularly interested in hidden costs. Labor and working time corresponded to 34% of the total expenditures in the open system (including blenderized diets), due to more prolonged manipulation and more frequent diet changes. More bags and given sets were consumed during such changes, representing 23% of the expenditures. Foodstuffs and nutrients did not exceed 43% of the financial estimates, thus confirming that labor and non-nutritional resources were responsible for most of the total costs. The closed system (ready to use products) was substantially cheaper, the difference between the two amounting to 46%²⁷.

Measurement of labor costs, including time required for each compounding step, is often a challenge. One example is home dietary counseling for patients discharged from the hospital. A survey with 1,100 professionals revealed that dietitians are typically engaged in such activity, and 2.8 hours are needed for each patient²⁷.

The information collected in the current protocol suggest that blenderized tube feeding diets are not necessarily cheaper than industrialized formulations, and would very likely exceed them, should indirect expenditures be calculated. Quality of nutritional support is another concern. In the study of Silkroski et al., the gap between prescribed and administered feedings tended to be larger when homebrew mixtures were adopted³⁴. Even in hospitals, failure to supply the full prescription is not uncommon, on account of gastrointestinal complications as well as frequent discontinuation of gavage because of diagnostic tests and other procedures^29-34.

In two investigations as little as 50% of the estimated volume was effectively supplied^30,34, a fact that could interfere with success of nutritional therapy, thus enhancing morbidity, mortality, length of hospitalization, and global healthcare costs². In case of outpatient care, the dangers are increased hospital admissions and higher needs for medical assistance during homecare³⁵. The inconsistencies in the composition of blenderized diets in this experience, which match those described by others^12,27, could compromise adequate nutrition with all the attending troubles of ongoing malnutrition.

 

Conclusion

Many reflections and discussions arise from the current study. Blenderized tube feeding diets were demonstrated to be highly variable and with inconsistent nutrient composition, two circumstances that hinder effective nutritional therapy. Preliminary cost assessment failed to confirm the widely announced benefit with regard to industrialized products.

Although not addressed in this protocol, other concerns related to BTD deserve to be mentioned such as excessive manipulation, increased risk of contamination, physical and chemical instability and higher osmolarity and viscosity. Labor and overhead are likely to be substantially higher, thus elevating total expenditures and reversing any theoretical savings. The consequences could also include more frequent complications and failure to correct malnutrition, thus perpetuating the disease.

This is a major challenge, and priority should be allocated to effective patient nutritional care and high quality of health services. Nutritional therapy demands quality control and patient monitoring at all times. Dietary products should be clearly labeled with regard to composition, origin and manipulation. Enteral access such as nasal tubes or ostomies deserves attention, along with administration routines and general clinical and biochemical monitoring²¹. Detailed protocols are recommended in order to curtail complications and improve the outcome of nutritional replenishment.

 

Acknowledgment

ILSI Brasil was founded in 1990 and it is one of the regional branches of the International Life Sciences Institute. It is a permanent forum for discussion and updating of knowledge in technical-scientific area, by joining efforts from scientists from academia, government, and industry. It currently has 34 member companies, including the main players in the food industry, acting through scientific committees and task forces. Its activities contribute to a better understanding of topics related to nutrition, food safety, toxicology, risk assessment, and the environment.

 

References

1. Waitzberg DL, Caiaffa WT, Correia MITD. Hospital Malnutrition: The Brazilian National Survey (IBRANUTRI): A Study of 4000 Patients. Nutrition 2001; 17: 573-80.         [ Links ]

2. Correia MITD, Waitzberg DL. The impact of malnutrition onmorbidity, mortality, length of hospital stay and costs evaluated through a multivariate model analysis. Clinical Nutrition 2003; 22 (3): 235-9.         [ Links ]

3. Barret JS, Shepaherd SJ, Gibson PR. Strategies to manage gastrointestinal symptoms complicating enteral feeding. JPEN 2009; 33: 21-6.         [ Links ]

4. Von Atzingen MCBC , Ribalta M, Santinho MAR, et al. Características físico químicas de dietas enterais artesanais com hidrolisado proteico de carne. Alim Nutr Araraquara abr/jun 2007; 18 (2): 183-9.         [ Links ]

5. Zaban ALRS, Novaes MRCG. Impact of the home enteral nutrition regulation issue in public hospitals in distrito federal, Brazil. e-SPEN. the European e-Journal of Clinical Nutrition and Metabolism 4 (2009) e193-e198.         [ Links ]

6. Lee CH, Hodgkiss IJ. The effect of poor handling procedures on enteral feeding systems in Hong Kong. Journal of Hospital Infection 1999; 42: 119-23.         [ Links ]

7. Okuma T, Nakamura M, Totake H, Fukumaga Y. Microbial Contamination of Enteral Feeding Formulas and Diarrhea. Nutrition 2000, 16 ( 9).         [ Links ]

8. Dudrick SJ, Palesty JA, Osigweh JM. 50 anos de terapia nutricional - do passado ao futuro. In: Waitzberg DL, Nutrição oral, enteral e parenteral na prática clínica. Atheneu - São Paulo. 2009. pp. 3-38.         [ Links ]

9. Hernndez JA, Torres NP, Jiménez M. Utilización clínica de la Nutrición Enteral. Nutr Hosp 2006; 21 (Supl. 2): 87-99.         [ Links ]

10. Araujo EM, Menezes HC. Estudo de fibras alimentares em frutas e hortaliças para uso em nutrição enteral ou oral. Ciênc. Tecnol. Aliment. (Internet). 2010 (cited 2013-01-14); 30(1): 42-47. Available from: http://www.scielo.br/pdf/cta/v30n1/aop_2946.pdf.         [ Links ]

11. Domene SMA, Galeazzi MAM. Prescrição e uso de formulados para nutrição enteral pelos serviços de nutrição hospitalares do município de campinas (SP). R. Nutr. PUCCAMP, Campinas, 1997 jul./dez,10 (2): 114-9.         [ Links ]

12. Mokhalalati JK, Druyan ME, Shott SB, Comer GM. Microbial, nutritional and physical quality of commercial and hospital prepared tube feedings in Saudi Arabia. Saudi Med J 2004; 25 (3): 331-41.         [ Links ]

13. Mitne C, Simões AMG, Wakamoto D, Liori GP, et al. Análise de dietas enterais artesanais. Rev Bras Nutr Clin 2001; 16 (3): 100-9.         [ Links ]

14. Costa GP, Silva MLT, Ferrini MT, et al. Estudo comparativo da contaminação microbiana das dietas enterais em sistema aberto e sistema fechado. Ver Bras Nutr Clin 1998; 13 (3): 180-8.         [ Links ]

15. Lima ARC, et al. Avaliação microbiològica de dietas enterais manipuladas em um hospital. Acta Cirúrgica Brasileira, 2005; S1: 27-30.         [ Links ]

16. Mauricio AA, Gazola S, Matioli G. Dietas enterais não industrializadas: análise microbiológica e verificação de boas práticas de preparação. Rev Nutr Campinas 2008; 21 (1): 29-37.         [ Links ]

17. The National Academies. Institute of Medicine. Dietary Reference Intakes (DRIs): Recommended Intakes for Individuals. Disponível em: www.iom.edu/File.aspx?ID=21372. acessado em 03/12/2012.         [ Links ]

18. Sullivan MM, Sorreda-Esguerra P, Platon MB, et al. Nutritional analysis of blenderized enteral diets in the Philippines. Asia Pac J Clin Nutr 2004; 13 (4): 385-90.         [ Links ]

19. Carvalho MLR, et al. Harzad analysis and critical control point system approach in the evaluation of environmental and procedural sources of contamination of enteral feedings in three hospitals. JPEN 2000; 24 (5).         [ Links ]

20. Rona MSS, Matioli G, Herrero F. Contaminação microbiana em dietas enterais artesanal, industrializada em pó e industrializada líquida. Rev Bras Nutr Clin 2005; 20 (3):111-16.         [ Links ]

21. Zamberlan P, Orlando PR, Dolce P, Delgado AF, Vaz FAC. Nutrição enteral em pediatria. Pediatria Moderna 2002; 8 (4): 105-24.         [ Links ]

22. Henriques GS, et al. Formulação de dietas enterais artesanais e determinação da osmolalidade pelo método crioscópico. Rev Nutr Campinas 1999; 12 (3): 225-32.         [ Links ]

23. Lochs H, Allison SP, et al. Introductory to the ESPEN Guidelines on Enteral Nutrition: Terminology, definitions and general topics. Clin Nutr 2006 Apr; 25 (2): 180-6.         [ Links ]

24. Freedland CP, Roller RD, Wolfe BM, Flynn NM. Microbial contamination of continuous drip feedings. J Parenter Enteral Nutr 1989; 13 (1): 18-22.         [ Links ]

25. Jalali M, et al. Bacterial contamination of hospital-prepared enteral tube feeding formulas in Isfahan, Iran. J Res Med Sci 2009 May; 14 (3): 149-56.         [ Links ]

26. Fernandez-Crehuet Navajas M, et al. Bacterial contamination of enteral feeds as a possible risk of nosocomial infection. J Hosp Infect 1992 Jun; 21 (2): 111-20.         [ Links ]

27. Silkroski M, et al. Tube Feeding Audit Reveals Hidden Costs and Risks of Current Practice. Nutr Clin Pract 1998; 13 (6): 283-90.         [ Links ]

28. Skipper A, Rotman N. A survey of the role of the dietitian in preparing patients for home enteral feeding. J Am Diet Assoc 1990 Jul; 90 (7): 939-44.         [ Links ]

29. Assis MCS, et al. Nutrição enteral: diferenças entre volume, calorias e proteínas prescritos e administrados em adultos. Rev Bras Ter Intensiva 2010; 22 (4): 346-50.         [ Links ]

30. O'Meara D, et al. Evaluation of delivery of enteral nutrition in critically ill patients receiving mechanical ventilation. Am J Crit Care 2008 Jan; 17 (1): 53-61.         [ Links ]

31. De Jonghe B, et al. A prospective survey of nutritional support practices in intensive care unit patients: what is prescribed? What is delivered? Crit Care Med 2001 Jan; 29 (1): 8-12.         [ Links ]

32. Campanella LCA, Silveira BM, et al. Terapia nutricional enteral: a dieta prescrita e realmente infundida? Rev Bras de Nutr Clin Braz J Clin Nutr 2008; 23 (1): 21-7.         [ Links ]

33. Teixeira ACC, Caruso L, Soriano FG. Terapia nutricional enteral em unidade de terapia intensiva: infusao versus necessidades. Rev Bras Ter Intensiva 2006; 18 (4): 331-7.         [ Links ]

34. McClave SA, Sexton LK, Spain DA, Adams JL, Owens NA, Sullins MB, et al. Enteral tube feeding in the intensive care unit: factors impeding adequate delivery. Crit Care Med 1999; 27 (7): 1252-6.         [ Links ]

35. Elia M, Stratton RJ, Russel C, et al. The cost of disease-related malnutrition in the UK and economic considerations for the use of oral nutritional supplements (ONS) in adults. A report by BAPEN. 2005.         [ Links ]

 

 

Correspondence:
Mariela Berezovsky
ILSI Brasil
Rua Hungria, 664 - conj. 113
01455-904 São Paulo, SP, Brazil
E-mail: mariela@ilsi.org.br /roseli.borghi@br.nestle.com

Recibido: 14-VIII-2013
Aceptado: 11-IX-2013