Associação das alterações do perfil glicêmico com o grau de processamento dos carboidratos: recorte de um ensaio clínico randomizado em pacientes pré-diabéticos

Abstract

Context: Type 2 diabetes mellitus (T2DM) is a chronic condition with high prevalence and growing impact. It is the most common type of diabetes, accounting for about 90% of cases. Prediabetes (pDM) is an asymptomatic and often underdiagnosed metabolic condition that precedes T2DM, reinforcing the importance of interventions at this stage. This study aims to assess the association between changes in glycemic profile after three months of intervention and the degree of carbohydrate processing. Methods: This is a subset analysis of the national multicenter randomized controlled clinical trial titled Diabetes Prevention Program: Pilot Study – PROVEN-Dia. The study sample consisted of 153 adult/older adult patients with pDM. Participants were randomly assigned to two treatment groups: the Usual Treatment Group (UTG), which received dietary counseling based on materials from the Brazilian Ministry of Health, and the PROVEN-Dia Group (PDG), which received guidance from a structured lifestyle-change program, including regular individual and group sessions, goal-setting, follow-up calls, and strategies to overcome barriers. Data were assessed before and after 3 months of intervention: anthropometry (weight, body mass index – BMI, and waist circumference – WC), dietary intake (two 24-hour dietary recalls at baseline and two at the end) with foods categorized according to the NOVA classification, and laboratory tests (insulin, fasting glucose, glycated hemoglobin – HbA1c, and Homeostatic Model Assessment for Insulin Resistance – HOMA-IR). Results: The majority of participants were female (74.5%), living with a partner (65.3%), non-white (74%), with a mean age of 49 years, belonging to social classes A and B (62.7%), and not using hypoglycemic medications (79.7%). The UTG reduced HbA1c, total carbohydrate intake in grams, and intake of carbohydrates from ultra-processed foods, whereas the PDG showed reductions in weight, BMI, WC, energy intake, total carbohydrate intake in grams, carbohydrate intake from ultra-processed foods and processed combined, carbohydrate intake from ultra-processed foods, and carbohydrate intake from unprocessed foods. A negative correlation was found between the change in HbA1c and the change in total carbohydrate intake in grams, indicating that as total carbohydrate consumption decreases, there is a corresponding reduction in HbA1c; however, this association was not observed for fasting glucose or HOMA-IR. Conclusions: The study revealed that after the interventions, there was a reduction in carbohydrate intake from all food sources, including unprocessed foods, with the greatest reduction observed for ultra-processed foods. Improvements were seen in cardiometabolic markers—reduction in body weight, BMI, and WC in the PDG—and HbA1c was associated with changes in total carbohydrate consumption after the intervention.