Introdução: O consumo de alimentos com elevado índice glicêmico e carga glicêmica tem sido associado ao aumento no risco de desenvolvimento de síndrome metabólica, importante precursor da doença cardiovascular e do diabetes mellitus tipo 2. Entretanto, esses achados ainda são inconsistentes e a utilização do índice glicêmico e da carga glicêmica para prevenção ou tratamento da síndrome metabólica e dos fatores de risco que a compõe ainda é controversa. Objetivos: Validar o diabetes mellitus autorreferido e verificar a associação do índice glicêmico e da carga glicêmica com a síndrome metabólica e seus componentes. Métodos: Foram utilizados dados provenientes do Inquérito de Saúde de São Paulo (ISA-Capital 2008) referentes a adultos e idosos de ambos os sexos residentes nessa cidade. Trata-se de estudo transversal, de base populacional, com amostra probabilística de indivíduos residentes em domicílios permanentes localizados na área urbana do município. As informações utilizadas são provenientes de um questionário estruturado, dois recordatórios alimentares de 24 horas, exames bioquímicos, valores aferidos de pressão arterial e medidas antropométricas (peso, estatura e circunferência da cintura). Foram estimadas as prevalências de diabetes mellitus e síndrome metabólica para o município de São Paulo. A validação diabetes mellitus autorreferido foi realizada mediante cálculo da sensibilidade, especificidade e valores preditivos positivo e negativo. O consumo alimentar habitual foi obtido por meio da incorporação dos dados alimentares no software Multiple Source Method. A associação entre índice glicêmico e carga glicêmica da dieta e síndrome metabólica e seus componentes foi verificada por meio de modelos de regressão logística estimados segundo faixa etária. Todas as análises levaram em consideração o desenho amostral do estudo. Resultados: As prevalências de diabetes mellitus e síndrome metabólica no município de São Paulo foram estimadas em 8,0 por cento e 30,2 por cento , respectivamente. A sensibilidade do diabetes mellitus autorreferido foi 63,8 por cento (IC 95 por cento : 49,2-76,3), a especificidade 99,7 por cento (IC 95 por cento : 99,1-99,9), o valor preditivo positivo 95,5 por cento (IC 95 por cento : 84,4-98,8) e o valor preditivo negativo 96,9 por cento (IC 95 por cento : 94,9-98,2). O índice glicêmico associou-se com a lipoproteína de alta densidade (OR: 1,16; IC 95 por cento : 1,02-1,32) em adultos e com a síndrome metabólica (OR: 1,24; IC 95 por cento : 1,1-1,37), a glicemia de jejum (OR: 1,15; IC 95 por cento : 1,01-1,31) e a pressão arterial (OR: 1,26; IC 95 por cento : 1,05-1,51) em idosos. Conclusão: O dado de diabetes mellitus autorreferido é válido, especialmente entre idosos residentes no município de São Paulo. Os resultados evidenciam a necessidade do rastreamento do diabetes mellitus em indivíduos assintomáticos que apresentem um ou mais fatores de risco para essa condição, principalmente na população adulta. No presente estudo, o IG da dieta associou-se à SM, glicemia de jejum e pressão arterial elevadas em idosos e apenas ao HDL-c baixo em adultos. As diferentes respostas entre os adultos e idosos podem sugerir que o índice glicêmico tem ação distinta entre os grupos etários. Ressalta-se que a qualidade do carboidrato parece ser mais importante do que a junção da qualidade-quantidade do carboidrato consumido para os parâmetros metabólicos avaliados na população da cidade São Paulo. / Introduction: High glycemic index and glycemic load intake has been associated with an increased risk for developing metabolic syndrome, an important precursor of cardiovascular disease and type 2 diabetes mellitus. However, these findings are inconsistent and the use of glycemic index and glycemic load for prevention or treatment of metabolic syndrome and the risk factors components is still controversial. Objectives: To validate self-reported diabetes mellitus and evaluate the association between glycemic index, glycemic load and metabolic syndrome and its components. Methods: Data were used from the Health Survey of São Paulo (ISA-Capital 2008) related to adults and elderly of both sexes living in this city. It is cross-sectional population-based study of individuals living in permanent homes located in the urban area of the municipality. Information used came from a structured questionnaire, two 24-hour dietary recalls, biochemical analysis, blood pressure and anthropometric measurements (weight, height and waist circumference). Prevalences of diabetes mellitus and metabolic syndrome were estimated for the city of São Paulo. The validation of self-reported diabetes mellitus was made by calculating the sensitivity, specificity, positive and negative predictive values. Usual food intake was achieved by the incorporation of food data in Multiple Source Method software. The association between glycemic index and glycemic load of the diet and metabolic syndrome and its components was verified by logistic regression models according to age group. All analysis took into account the sampling design of the study. Results: Diabetes mellitus and metabolic syndrome prevalences in São Paulo city were 8.0 per cent and 30.2 per cent , respectively. The sensitivity of self-reported diabetes mellitus was 63.8 per cent (95 per cent CI: 49.2 to 76.3), specificity was 99.7 per cent (95 per cent CI: 99.1 to 99.9), the positive predictive value was 95.5 per cent (95 per cent CI: 84.4 to 98.8) and the negative predictive value was 96.9 per cent (95 per cent CI: 94.9 to 98.2). Glycemic index was associated with high density lipoprotein cholesterol (OR: 1.16; 95 per cent CI: 1.02 to 1.32) in adults and with metabolic syndrome (OR: 1.24; 95 per cent CI: 1.1 to 1, 37), fasting blood glucose (OR: 1.15; 95 per cent CI: 1.01 to 1.31) and blood pressure (OR: 1.26; 95 per cent CI: 1.05 to 1.51) in elderly. Conclusion: Self- report diabetes mellitus data is valid, especially among elderly people living in São Paulo. The results show the need for diabetes mellitus screening in asymptomatic individuals who have one or more risk factors for this condition, especially in adults. Glycemic index was associated with metabolic syndrome and elevated fasting blood glucose and blood pressure in elderly and only with low high density lipoprotein cholesterol in adults. The different responses among adults and elderly may suggest that glycemic index has distinct action between age groups. Carbohydrate quality seems to be more important than the joint quality-quantity of the ingested carbohydrate for metabolic parameters evaluated in the population of São Paulo city.

Phenolic compounds have attracted much attention due to numerous health benefits, including high antioxidant properties, reduced risk of cancer, and inhibition of digestive enzymes. Recent research has suggested that different phenolic compounds may interact with starch. The first objective was to investigate the effect of green or black tea extracts on hydrolysis of wheat, rice, corn, and potato starches. Cooking starches in the presence of either tea reduced their hydrolysis. Potato starch cooked with black tea was the most effective treatment. Observations suggested that hydrolysis may be affected by interactions and by impact on specific enzymes based on starch structure. The second objective was to determine if similar effect could be observed in product system. Addition of green tea extract to sponge cake significantly reduced in vitro starch digestibility, thus could reduce the expected glycemic index. In addition, significant increases in dietary fibre, resistant starch, and antioxidant properties were observed.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal

Thesis (Ph.D. in Epidemiology, Dept. of Preventive Medicine and Biometrics) -- University of Colorado Denver, 2008. / Typescript. Includes bibliographical references (leaves 81-92). Free to UCD Anschutz Medical Campus. Online version available via ProQuest Digital Dissertations;

Objetivos. Investigar a associação entre consumo alimentar, índice glicêmico (IG) da dieta habitual com glicemia e insulinemia de jejum, resistência à insulina (HOMA R) e alteração da homeostase glicêmica (AHG: glicemia de jejum alterada - GJA, tolerância à glicose diminuída - TGD e diabetes mellitus tipo 2). Indivíduos e métodos. 1.054 nipo-brasileiros residentes em Bauru, ambos os sexos, 30 a 90 anos, primeira e segunda gerações, que participaram de inquérito transversal conduzido em 2000, sem diagnóstico prévio de AHG e/ou uso de hipoglicemiantes orais/insulina. Para avaliação da dieta habitual, utilizou-se questionário quantitativo de freqüência alimentar, previamente validado, com o auxílio do programa Dietsys versão 4.0. Os fatores dietéticos foram ajustados pelas calorias totais pelo método residual, após transformação logarítmica. As variáveis independentes foram inicialmente selecionadas segundo correlação de Pearson com glicemia e insulinemia de jejum ou HOMA R (variáveis dependentes contínuas) com valor de p<0,20. A associação entre nutrientes/ alimentos com as variáveis dependentes contínuas foi avaliada em modelos de regressão linear múltiplos. Modelos de regressão logística múltiplos foram utilizados para investigar a relação entre fatores dietéticos (tercis) com AHG, ajustados por fatores de confusão. Resultados. Após ajuste múltiplo, houve associações inversas entre o consumo dos grupos de laticínios integrais (g/dia), doces (g/dia), IG da dieta e glicemia de jejum (mg/dl). Relação inversa entre o consumo de IG da dieta e HOMA R também foi observada. Verificou-se razão de chances - RC (IC 95%) - para GJA de 1,70 (1,06 - 2,74) no último tercil de consumo de fibras totais em que as principais fontes alimentares foram o arroz polido, pão branco e frutas/sucos de frutas. Efeito protetor entre o tercil intermediário do consumo de vegetais [0,61 (0,38 - 0,98)] foi observado para TGD. Conclusão. Em nossa população de estudo, a fibra dietética proveniente do consumo excessivo de arroz polido, pão branco, frutas e sucos representou risco para AHG. Por outro lado, observou-se efeito protetor do maior consumo de vegetais para TGD. / Objective. To investigate the effects of food intake and dietary glycemic index (GI) on fasting plasma levels of glucose and insulin, homeostasis model assessment of insulin resistance (HOMA R) and impaired glucose disturbance - IGD (impaired fasting glucose - IFG, impaired glucose tolerance - IGT and diabetes mellitus type 2). Subjects and methods. 1,054 Japanese-Brazilians living in Bauru, of both genders, 30 to 90 years-old, first- and second-generation, who completed all the information for the cross-sectional survey in 2000, without previous diagnosis of IGD and/or use of oral hipoglycemic medication/insulin. Food consumption was assessed using a validated food frequency questionnaire, with the aid of the software Dietsys 4.0. All dietary factors were log-transformed and adjusted for total energy intake by residual method. The dependent variables were first selected using Pearson correlation with fasting plasma levels of glucose and insulin or HOMA R (continuous independent variables), with p<0.20. The associations between selected nutrients/foods and the continuous independent variables were assessed using multiple linear regression models. Logistic regression models were used to investigate the relationship between dietary factors (tercile) and IGD, while adjusting for confounding factors. Results. After multiple adjustments, intakes of whole dairy products and sweets (g/day) and dietary GI were inversely correlated with fasting glucose levels (mg/dl). Dietary GI was also inversely correlated with HOMA R. The odds ratio (95% confidence interval) for IFG was 1.70 (1.06 - 2.74) in the highest tercile of total dietary fiber (main food sources: white rice and bread, fruits/fruit juices). The second tercile of vegetable intake was associated with a risk reduction of IGT [0.61 (0.38 - 0.98)]. Conclusion. In our study population, the intake of total dietary fiber, largely attributable to high intakes of white rice, white bread, fruits/fruit juices, was positively associated with risk of IGD. On the other hand, a protective effect was observed for higher intake levels of vegetables

La réduction des excursions glycémiques postprandiales a été proposée comme un moyen pour limiter le risque de développement du diabète de type 2. L’intérêt s’est donc porté sur les outils nutritionnels susceptibles de moduler la biodisponibilité des glucides et ainsi leur impact sur la glycémie postprandiale. Les travaux réalisés au cours de cette thèse avaient pour but d’étudier les effets de différents ingrédients modifiant la biodisponibilité du glucose, non seulement sur la glycémie postprandiale à court terme (2 heures) mais aussi sur les cinétiques du débit d’apparition et de disparition de glucose (total, exogène et endogène - isotopes stables) et les autres paramètres métaboliques de la phase postprandiale au cours de la journée. Dans la première étude (β-glucanes), nous avons montré que l’addition de fibres β-glucanes à un repas glucidique chez des sujets sains en surpoids ralentit l’absorption du glucose dans le plasma. Ceci a prolongé la réponse insulinique et par conséquent l’inhibition de la lipolyse et de la production endogène de glucose. Dans la deuxième étude (Eurostarch), nous avons montré que la diminution de la biodisponibilité du glucose au petit-déjeuner (amidon lentement digestible, index glycémique bas) diminue l’apparition du glucose exogène dans le plasma et pourrait avoir un effet second-repas chez des sujets sains en surpoids. Mais nous n’avons pas mis en évidence d’amélioration de ces effets métaboliques à plus long terme (5 semaines). Dans la troisième étude présentée (Nutriose), nous avons montré que l’addition de dextrine résistante NUTRIOSE®10 (fermentescible) au petit-déjeuner chez des sujets sains, diminue les réponses glycémiques, insuliniques postprandiales et le profil de ghréline au cours de la journée (en comparaison à une maltodextrine). En parallèle, la prolongation observée de la fermentation et l’oxydation du NUTRIOSE®10 pourraient fournir de l’énergie en phase postprandiale tardive. En conclusion, l’analyse des paramètres métaboliques au-delà de 2 heures après le repas, a permis de mettre en évidence les effets métaboliques à plus long terme de la modulation de l’apparition du glucose dans le plasma (ralentissement, prolongation, réduction) sur les cinétiques du glucose, la réponse insulinique, la lipolyse et l’oxydation des substrats / The reduction of the postprandial glycemic excursions has been proposed as to limit risk of type 2 diabetes. There has been growing interest in the development of dietary ingredients that could potentially modulate carbohydrates bioavailability and thus their impact on postprandial glycemia. The aim of this thesis was to investigate the effects of the the modulation of glucose bioavailability by different ingredients on 2-hour glycemic response but also on glucose kinetics (total, exogenous and endogenous – stable isotopes) and on other daylong metabolic parameters. In the first study (β-glucanes), we showed that the addition of β-glucan fiber to a carbohydrate meal in healthy overweight subjects reduced the appearance of glucose in plasma. As a consequence, insulin response was also prolonged and induced a prolonged inhibition on lipolysis and endogenous glucose production. In the second study (eurostarch), the reduction in glucose availability (slowly available glucose, low GI) at breakfast decreased plasma exogenous glucose appearance and tended to improve glucose control at the subsequent lunch. But we did not observe the improvement of such metabolic effects in the long-term (5 weeks). In the last study, we showed that the addition of a resistant dextrin, NUTRIOSE®10, decreased postprandial glycemic and insulinemic response as well as daylong satiety-related ghrelin profile, compared to maltodextrin. In parallel, the prolonged fermentation and oxidation pattern of NUTRIOSE®10 up to 10 hours after ingestion at breakfast could induced an extended energy release with NUTRIOSE®10 in the late postprandial phase. In conclusion, the follow-up of metabolic parameters beyond 2 hours after the meal have highlighted the longer-term metabolic effects of the modulation of glucose appearance in plasma (delay, extension, reduction) on glucose kinetics, insulin response, lipolysis and nutrients oxidation

No description available.

A hipertensão é uma das principais causas de morbidade e mortalidade no Brasil. Os hipertensos muitas vezes apresentam perfil lipídico e glicidico desfavoráveis. A alimentação pode desempenhar um papel importante na redução da pressão arterial (PA) e no perfil lipídico e controle glicêmico desses pacientes. Avaliar o impacto de uma intervenção nutricional adaptada ao padrão alimentar brasileiro no controle dos níveis pressóricos e metabólico de pacientes hipertensos em acompanhamento em um serviço de atenção primária de saúde do município de São Luís do Maranhão. Metodologia: ensaio clínico randomizado utilizando uma dieta de baixo índice glicêmico combinada ao aumento do consumo de frutas, vegetais, grãos integrais e laticínios desnatados que são os princípios do Dietary Approach to Stop Hypertension (dieta DASH). Foram alocados randomicamente 206 pacientes hipertensos que foram acompanhados por 6 meses. O grupo controle (GC, n=101) recebeu aconselhamento padrão, focado na redução da ingestão de sal. Resultados: Dos 206 pacientes randomizados, 156 (37 homens, 119 mulheres) completaram o estudo. A idade média dos participantes foi de 60,1 (DP 12,9) anos. Após 6 meses, houve redução na média da pressão arterial sistólica (PAS) em 14,4 mmHg e na diastólica (PAD) de 9,7 mmHg no grupo experimental (GE), em comparação a 6,7 mmHg e 4,6 mmHg, respectivamente, no GC. Após o ajuste para mudança de peso corporal, PA na linha de base e idade, essas diferenças entre os grupos foram de aproximadamente 9,2 mmHg e 6,2 mmHg, respectivamente. Ocorreram tambem variações estatisticamente significantes na excreção urinária de sódio, reduzida em 43,4 mEq/24 h no GE, bem como o colesterol total (-46.6mg/dl) , LDL colesterol (-42.5mg/dl), triglicérides (-31.3mg/dl), glicemia de jejum (-9.6mg/dl ) e hemoglobina glicada (-0,1%). O consumo alimentar modificou-se no GE com aumento do consumo de vegetais, passando de 2,97 para 5,85 ; frutas (4,09-7,18); feijão (1,94-3,13) e peixes (1,80 para 2,74). Modificações importantes relacionadas à redução significativa de carboidratos, teor lipídico e carga glicêmica da dieta, foram observadas. Conclusão: Este estudo mostrou a viabilidade e a eficácia de uma abordagem dietética com base no padrão alimentar brasileiro, na redução da PA e parâmetros bioquímicos inadequados, podendo causar um grande impacto na saúde pública. / Hypertension is one of the leading causes of morbidity and mortality in Brazil. Hypertensive patients often have unfavorable lipid profile and glucose level. Nutrition may play an important role in reducing blood pressure (BP) and metabolic control of these patients. Objective: To evaluate the impact of nutritional intervention adapted to the Brazilian food in controlling blood pressure and metabolic monitoring in hypertensive patients from a primary care service in São Luís do Maranhão. Methodology: Randomized clinical trial using a low-glycemic index diet combined with increased consumption of fruits, vegetables, whole grains and nonfat dairy products which principles of the Dietary Approach to Stop Hypertension (DASH). We randomly assigned 206 patients with hypertension who were followed for 6 months. The control group (CG, n = 101) received standard counseling, focused on reducing salt intake. Results: Of the 206 patients randomized, 156 (37 men, 119 women) completed the study. The average age of participants was 60.1 (SD 12.9) years. After 6 months, a reduction in mean systolic blood pressure (SBP) by 14.4 mmHg and diastolic blood pressure (DBP) by 9.7 mmHg in the experimental group (EG), compared to 6.7 mmHg and 4.6 mmHg, respectively GC. After adjusting for change in body weight at baseline BP and age, these differences between groups were approximately 9,2 mmHg and 6,2 mmHg, respectively. There were also statistically significant variability in urinary sodium excretion, reduced by 43.4 mEq/24 h at GE, as well as total cholesterol (-46.6mg/dl), LDL cholesterol (-42.5mg/dl), triglycerides (-31.3 mg / dl), fasting glucose (-9.6mg/dl) and glycated hemoglobin (-0.1%). The EG increased the intake of vegetables, from 2.97 to 5.85; fruits (4,09 to 7.18), beans (1.94 to 3.13) and fish (1.80 to 2.74).Significant changes related to the significant reduction of carbohydrate, lipid content and glycemic load of the diet have been observed. Conclusion: This study showed the feasibility and effectiveness of a dietary approach based no Brazilian pattern in reducing blood pressure and biochemical parameters. Dietary changes as proposed may have a great impact on public health.

A hipertensão é uma das principais causas de morbidade e mortalidade no Brasil. Os hipertensos muitas vezes apresentam perfil lipídico e glicidico desfavoráveis. A alimentação pode desempenhar um papel importante na redução da pressão arterial (PA) e no perfil lipídico e controle glicêmico desses pacientes. Avaliar o impacto de uma intervenção nutricional adaptada ao padrão alimentar brasileiro no controle dos níveis pressóricos e metabólico de pacientes hipertensos em acompanhamento em um serviço de atenção primária de saúde do município de São Luís do Maranhão. Metodologia: ensaio clínico randomizado utilizando uma dieta de baixo índice glicêmico combinada ao aumento do consumo de frutas, vegetais, grãos integrais e laticínios desnatados que são os princípios do Dietary Approach to Stop Hypertension (dieta DASH). Foram alocados randomicamente 206 pacientes hipertensos que foram acompanhados por 6 meses. O grupo controle (GC, n=101) recebeu aconselhamento padrão, focado na redução da ingestão de sal. Resultados: Dos 206 pacientes randomizados, 156 (37 homens, 119 mulheres) completaram o estudo. A idade média dos participantes foi de 60,1 (DP 12,9) anos. Após 6 meses, houve redução na média da pressão arterial sistólica (PAS) em 14,4 mmHg e na diastólica (PAD) de 9,7 mmHg no grupo experimental (GE), em comparação a 6,7 mmHg e 4,6 mmHg, respectivamente, no GC. Após o ajuste para mudança de peso corporal, PA na linha de base e idade, essas diferenças entre os grupos foram de aproximadamente 9,2 mmHg e 6,2 mmHg, respectivamente. Ocorreram tambem variações estatisticamente significantes na excreção urinária de sódio, reduzida em 43,4 mEq/24 h no GE, bem como o colesterol total (-46.6mg/dl) , LDL colesterol (-42.5mg/dl), triglicérides (-31.3mg/dl), glicemia de jejum (-9.6mg/dl ) e hemoglobina glicada (-0,1%). O consumo alimentar modificou-se no GE com aumento do consumo de vegetais, passando de 2,97 para 5,85 ; frutas (4,09-7,18); feijão (1,94-3,13) e peixes (1,80 para 2,74). Modificações importantes relacionadas à redução significativa de carboidratos, teor lipídico e carga glicêmica da dieta, foram observadas. Conclusão: Este estudo mostrou a viabilidade e a eficácia de uma abordagem dietética com base no padrão alimentar brasileiro, na redução da PA e parâmetros bioquímicos inadequados, podendo causar um grande impacto na saúde pública. / Hypertension is one of the leading causes of morbidity and mortality in Brazil. Hypertensive patients often have unfavorable lipid profile and glucose level. Nutrition may play an important role in reducing blood pressure (BP) and metabolic control of these patients. Objective: To evaluate the impact of nutritional intervention adapted to the Brazilian food in controlling blood pressure and metabolic monitoring in hypertensive patients from a primary care service in São Luís do Maranhão. Methodology: Randomized clinical trial using a low-glycemic index diet combined with increased consumption of fruits, vegetables, whole grains and nonfat dairy products which principles of the Dietary Approach to Stop Hypertension (DASH). We randomly assigned 206 patients with hypertension who were followed for 6 months. The control group (CG, n = 101) received standard counseling, focused on reducing salt intake. Results: Of the 206 patients randomized, 156 (37 men, 119 women) completed the study. The average age of participants was 60.1 (SD 12.9) years. After 6 months, a reduction in mean systolic blood pressure (SBP) by 14.4 mmHg and diastolic blood pressure (DBP) by 9.7 mmHg in the experimental group (EG), compared to 6.7 mmHg and 4.6 mmHg, respectively GC. After adjusting for change in body weight at baseline BP and age, these differences between groups were approximately 9,2 mmHg and 6,2 mmHg, respectively. There were also statistically significant variability in urinary sodium excretion, reduced by 43.4 mEq/24 h at GE, as well as total cholesterol (-46.6mg/dl), LDL cholesterol (-42.5mg/dl), triglycerides (-31.3 mg / dl), fasting glucose (-9.6mg/dl) and glycated hemoglobin (-0.1%). The EG increased the intake of vegetables, from 2.97 to 5.85; fruits (4,09 to 7.18), beans (1.94 to 3.13) and fish (1.80 to 2.74).Significant changes related to the significant reduction of carbohydrate, lipid content and glycemic load of the diet have been observed. Conclusion: This study showed the feasibility and effectiveness of a dietary approach based no Brazilian pattern in reducing blood pressure and biochemical parameters. Dietary changes as proposed may have a great impact on public health.

The ability of the soluble fibre (1->3)(1->4)-β-D-glucan to attenuate postprandial glycemic responses depends on its viscosity which, in turn, depends on molecular weight (MW) and dose. However, the effect of altering viscosity by changing solution volume is unknown. Furthermore, β-glucan solutions may form hard gels when left to age, but the effect of these gels on glycemic responses is unknown. Therefore, the effects of varying the MW and volume of β-glucan solutions and hard gels, on glycemic responses were determined. The results showed that glycemic responses were reduced by increasing viscosity by increasing MW but not by reducing solution volume. Although β-glucan gels reduced the rate of glucose diffusion in vitro, they had no effect on glycemic responses in vivo. Thus, changing solution viscosity through changes in volume does not alter the effect of β-glucan on glycemic response, and β-glucan gels are ineffective at attenuating in vivo glycemic responses.