Desenvolvimento e validação de métodos analíticos para avaliação dos parâmetros de qualidade de kombuchas comerciais brasileiras

Abstract

The growing interest in healthy habits and so-called functional foods has driven the consumption of Kombucha, a fermented beverage made from the Camellia sinensis plant, sugar, and a symbiotic consortium of bacteria and yeasts (SCOBY). Originally from China, Kombucha has gained global market presence and, in 2024, generated approximately 2.97 billion dollars, with continued growth projected. This scenario highlights the importance of quality control in the production of this beverage, aiming to ensure consumer safety, compliance with regulations, and consistency of the final product. In this context, the present study aimed to develop and validate analytical methods applicable to the quality control of commercial Brazilian Kombuchas. A wide variety of samples from 30 different brands from various regions of Brazil were analyzed, providing a comprehensive overview of the composition of products available in the market. The samples had a declared alcohol content of up to 0.5% v/v (classified as non-alcoholic). A method was developed and validated for the simultaneous determination of alcohol content and volatile acidity in Kombuchas using gas chromatography with flame ionization detection (GC-FID). Ethanol levels ranged from 0.10 to 1.70% v/v, with 60% of the samples exceeding the legal limit for non-alcoholic beverages. Volatile acidity ranged from 17 to 71 mEq/L, with half of the samples falling outside the recommended range (30 to 130 mEq/L). Total sugar content was quantified by UV-VIS spectrophotometry using the DNS reagent, showing variations between 20 g/L and 84 g/L. A method was also developed to evaluate the sugar and sweetener profile (whose addition is not permitted in Kombuchas in Brazil) using gas chromatography coupled with mass spectrometry (GC-MS). None of the investigated sweeteners (erythritol, xylitol, or sucralose) were detected in any of the samples. Glucose levels ranged from 2.5 to 23.9 g/L, while fructose ranged from 2.5 to 29.8 g/L. Sucrose was quantified in 28 samples, with concentrations from 3.6 to 29.9 g/L, while the remaining samples were above the method’s working range. Additionally, chemometric models based on mid-infrared (MIR) spectroscopy and partial least squares (PLS) regression were developed and validated for the quantification of alcohol content, volatile acidity, and sugars. The models demonstrated good performance, with RPD values above 2.4 and low root mean square error of prediction (RMSEP), enabling rapid analyses with minimal sample preparation. Validation confirmed the linearity and precision of the models, with deviations below 5%. The results obtained in this study revealed inconsistencies between labeling information and the actual composition of the analyzed Kombucha samples, underscoring the need for stricter regulatory oversight. The developed and validated methods proved to be reliable and efficient alternatives for evaluating compliance with current legislation and may offer significant contributions to regulatory agencies.