Skip to main content
Download PDF
Download ePub

The link between ultra-processed food consumption, fecal microbiota, and metabolomic profiles in older mediterranean adults at high cardiovascular risk

Nutrition Journal volume 24, Article number: 62 (2025) Cite this article

Abstract

Background

Ultra-processed food (UPF) consumption has been linked to adverse metabolic outcomes, potentially mediated by alterations in gut microbiota and metabolite production.

Objective

This study aims to explore the cross-sectional and longitudinal associations between NOVA-classified UPF consumption, fecal microbiota, and fecal metabolome in a population of Mediterranean older adults at high cardiovascular risk.

Methods

A total of 385 individuals, aged between 55 and 75 years, were included in the study. Dietary and lifestyle information, anthropometric measurements, and stool samples were collected at baseline and after 1-year follow-up. Fecal microbiota and metabolome were assessed using 16 S rRNA sequencing and liquid chromatography-tandem mass spectrometry, respectively.

Results

At baseline, higher UPF consumption was associated with lower abundance of Ruminococcaceae incertae sedis (β = − 0.275, P = 0.047) and lower concentrations of the metabolites propionylcarnitine (β = − 0.0003, P = 0.013) and pipecolic acid (β = − 0.0003, P = 0.040) in feces. Longitudinally, increased UPF consumption was linked to reduced abundance of Parabacteroides spp. after a 1-year follow-up (β = − 0.278, P = 0.002).

Conclusions

High UPF consumption was associated with less favorable gut microbiota and metabolite profiles, suggesting a possible link to reduced short-chain fatty acid (SCFA) production, altered mitochondrial energy metabolism, and impaired amino acid metabolism. These findings support the reduction of UPF consumption and the promotion of dietary patterns rich in fiber for better gut health. Further research is needed to confirm these associations and clarify the underlying mechanisms.

Trial registration

: ISRCTN89898870 (https://doiorg.publicaciones.saludcastillayleon.es/10.1186/ISRCTN89898870).

Peer Review reports

Introduction

Ultra-processed foods (UPF), categorized as Group 4 in the NOVA food classification system [1], are industrial formulations primarily composed of substances derived from foods, along with additives and cosmetic ingredients. These products typically contain minimal whole-food content and are designed for convenience, enhanced palatability, and long shelf life [2].

The global increase in UPF consumption has emerged as an important public health concern, as it has been linked to several adverse health outcomes [3], including obesity [4], cardiovascular disease, and metabolic disorders [5, 6]. To understand the impact of UPF consumption on human physiology and metabolism, nutrition research has primarily focused on the dietary characteristics of UPF, such as energy density and the content of added fats, sugars, and salt. However, other aspects introduced during food processing could play an equally substantial role, particularly by triggering inflammation-related processes through interactions between the diet, microbiome, and host [7].

The gut microbiota, a complex ecosystem influenced by dietary and lifestyle factors, has emerged as a potential mediator of diet-related health outcomes [7,8,9,10]. Recent evidence suggests that UPF consumption may negatively impact the diversity and abundance of beneficial bacteria, potentially disrupting metabolic pathways crucial for health [11, 12]. In contrast, dietary patterns like the Mediterranean diet (MedDiet), characterized by a higher intake of whole and minimally processed foods, have been shown to induce changes in gut microbiota composition linked to better metabolic health [13, 14].

Despite these associations, the relationship between UPF consumption and gut microbiota remains insufficiently explored, particularly in the context of long-term dietary changes. A previous study from our group suggested an association between gastrointestinal inflammation-related taxa and high UPF consumption in a cohort of older adults at high cardiovascular risk. However, this study only demonstrated cross-sectional associations [12]. A recent review article highlights the limited availability of research and the need for further human and animal studies to better understand the effect of UPF on the gut microbiome [15]. To address this research gap, our study aims to provide new insights into the complex interplay between diet, gut microbiota, and metabolic health by examining how baseline and long-term changes in UPF consumption influence fecal microbial composition and metabolic profiles.

Utilizing the large-scale randomized clinical trial PREDIMED (PREvención con DIeta MEDiterránea)-Plus [16] as a platform, and integrating 16 S rRNA sequencing and metabolomics, we explore both cross-sectional and longitudinal associations between NOVA-classified UPF consumption, fecal microbiota composition, and fecal metabolome in a cohort of 385 individuals. We hypothesize that higher UPF consumption may be associated with a less favorable gut microbiota composition and metabolic profiles, both at baseline and after one year of follow-up. The study offers the opportunity to explore how diets rich in UPF might influence gut health.

Methods

Study design and participants

This study was conducted as part of the PREDIMED-Plus randomized clinical trial, using an observational cross-sectional and longitudinal design.

The PREDIMED-Plus trial is a 6-year, multicenter, parallel-group, randomized, single-blind intervention designed to evaluate the long-term effects of a lifestyle intervention—including an energy-reduced Mediterranean diet, physical activity promotion, and behavioral support for weight loss—compared with a traditional Mediterranean diet with ad libitum caloric intake, on cardiovascular disease and mortality. Eligible participants were men and women aged between 55 and 75 years without cardiovascular disease at baseline, with a baseline body mass index (BMI) between 27 and 40 kg/m², who met at least three criteria for metabolic syndrome [17]. Participants were randomized in a 1:1 ratio to either the intervention or control group. Additional details are provided elsewhere [16].

Detailed information regarding participant selection is available elsewhere [14]. A subsample of 400 participants, recruited from centers in Alicante, Barcelona, Reus, and Valencia, had fecal microbiota 16 S rRNA sequencing and fecal metabolomics data available at both baseline and after a 1-year follow-up. Of these 400 participants, 15 were excluded since they were outside the pre-defined range for total energy intake.

General assessments and anthropometric measurements

Participants completed a comprehensive questionnaire to collect data on sociodemographic and medical history. Leisure-time physical activity was estimated using the validated REGICOR Short Physical Activity Questionnaire [18]. Waist circumference was measured twice at the midpoint between the lowest rib and the iliac crest using an anthropometric measuring tape. Body weight and height were also measured twice, using calibrated electronic scales and a wall-mounted stadiometer, respectively. The mean of both measurements was calculated and used.

Dietary assessments

Dietary intake was assessed by trained dietitians using a validated 143-item semi-quantitative Food Frequency Questionnaire (FFQ) [19] during face-to-face visits at baseline and at the 1-year follow-up. Detailed information is available elsewhere [16]. Nutrient intake, including sodium, saturated and trans fatty acids, fiber, alcohol (g/day), and total energy (kcal/day), was calculated using Spanish food composition Table [20]. The intake of specific food groups, such as fruits and vegetables (g/day), was also determined.

Participants who did not complete the FFQ at baseline or after the 1-year follow-up, or whose total energy intake fell outside the pre-specified limits (women: <500 or > 3500 kcal/day; men: <800 or > 4000 kcal/day), were excluded from the analysis (N = 15) [21].

The NOVA Food Classification system [2] was used to categorize food and beverage items from the FFQ into four groups based on their degree of processing. Using the FFQ data, we calculated the intake of foods and beverages in the different NOVA groups, expressed in g/day. In the present study, we focused on food items classified as NOVA Group 4, which includes ultra-processed food and drink products, considered as a continuous variable.

Stool sample collection

Participants were instructed to collect stool samples and to keep them frozen until delivery to the laboratory within 12 h after excretion, as previously reported [14]. Specifically, participants were provided with a stool collection kit containing a disposable cardboard urinal, a 50 mL sterile container with a spatula, a portable cooler, cold packs, and a sealed plastic container. Detailed instructions were given to ensure proper sample collection, storage, and transport. Participants collected a stool sample (approximately the size of a walnut) using the provided spatula, sealed the container, and immediately placed it in their home freezer. On the day of delivery, the sample was transported in the portable cooler with frozen cold packs to maintain a frozen state. Upon receipt, researchers processed the samples in a refrigerated environment, dividing each sample into 250 mg aliquots, which were stored in labeled cryovials at − 80 °C until analysis.

Fecal metabolomics analyses

Metabolomic profiling of stool samples collected at baseline and at the 1-year follow-up was conducted using a liquid chromatography-tandem mass spectrometry platform. Detailed information is available elsewhere [14]. The fecal metabolomics data included a total of 532 identified metabolites.

Fecal bacterial DNA extraction and 16 S amplicon sequencing

Microbial DNA was extracted using the QIAamp PowerFecal DNA Kit (Qiagen, Hilden, Germany), and DNA quality was assessed with the Qubit 2.0 Fluorometer (Thermo Fisher Scientific, Waltham, MA, USA). The V4 region of the 16 S rRNA gene was amplified in triplicate PCR reactions, followed by purification and quantification of the PCR products. Sequencing libraries were constructed from purified PCR products and sequenced on an Illumina NovaSeq platform, with mock communities and negative controls included for quality assurance. Detailed information on these analytical steps can be found elsewhere [14].

Amplicon sequence variants (ASV) were determined using the DADA2 pipeline [22], and taxonomy was assigned based on the Silva database [23], version 138.1. A total of 218 genera were identified after filtering the raw ASV counts to include only features with a total relative abundance ≥ 0.001 in at least 10% of the samples.

Statistical analyses

All statistical analyses were conducted using the R programming language and software environment (version 4.4.1) along with the RStudio integrated development environment (version 2023.6.0.421).

The baseline characteristics of the study population and 1-year changes were summarized using means and standard deviations for continuous variables and using numbers and percentages for categorical variables. Differences between timepoints for continuous variables were tested using paired t-tests. UPF consumption (in g/day) was adjusted for total energy intake using the residual regression method [24]. This approach allows UPF consumption to be expressed in g/day rather than as a percentage of energy intake, to account for foods with little or no caloric content (e.g., artificially sweetened beverages) and considering non-nutritional factors related to food processing (e.g., presence of food additives).

Fecal microbiota alpha diversity indices, including Chao1, Shannon, and Simpson [25,26,27], were calculated from absolute ASV counts. For the cross-sectional assessment, the association between calculated indices and baseline UPF consumption was examined by fitting linear regression models and adjusting for selected covariates: age, sex, center, smoking status, education level, prevalence of diabetes, hypertension and hypercholesterolemia, BMI, physical activity, alcohol intake, fiber intake, and consumption of food and beverages categorized by other NOVA groups. Before fitting the linear models, numerical covariates were standardized using z-scores. In a similar manner, for the longitudinal assessment, we explored the association between changes in alpha diversity and changes in the UPF consumption over one year. These changes were calculated by subtracting baseline values from the values at 1 year. The PREDIMED-Plus study arm was also included as a covariate. Results with P < 0.05 were reported as significant.

Beta diversity was evaluated by calculating Euclidean distance on centered log-ratio transformed genus counts (Aitchison distance) [28], including only features with a total relative abundance of ≥ 0.001 in at least 10% of observations. Permutational multivariate analysis of variance (PERMANOVA) [29] was implemented using the “adonis2” function from the R package vegan (version 2.6–6.1) (available at https://CRAN.R-project.org/package=vegan) to test whether differences in fecal microbiota composition were significantly associated with the given predictors. Results with p < 0.05 were reported as significant.

For the cross-sectional analysis, the association between differences in Aitchison distance and UPF consumption at baseline was tested, accounting for covariates. For the longitudinal analysis, the association between differences in Aitchison distance, calculated between baseline and 1-year counts, and the interaction term UPF 1-year change × time was tested. In this analysis, participant IDs were specified as a variable within which permutations were constrained, and the PREDIMED-Plus study arm of the trial was included as an additional covariable.

Differential abundance analysis was performed using the MaAsLin2 R package [30], utilizing the same input data as the beta diversity analysis. General linear models were fitted, adjusting for the same covariates considered in the alpha and beta diversity association analyses, with the total number of reads included as an additional covariate. No further filtering was applied in the model. Continuous covariates were standardized to the same scale. Results were subset based on the exposure of interest, and Benjamini-Hochberg adjusted P-values were calculated, focusing comparisons solely on the predictor of interest while controlling for covariates. Findings with adjusted P-value < 0.05 were reported.

In the cross-sectional assessment, the association between fecal microbiota genera abundance and UPF consumption at baseline was tested. For the longitudinal assessment, the association between the 1-year change in fecal microbiota genera abundance and the 1-year change in UPF consumption was evaluated. The time between baseline and follow-up, as well as the PREDIMED-Plus study group, was included as fixed effects along with the other covariates. Participant IDs were specified as random effects to account for the non-independence of samples from the same individual.

Output files generated with DADA2 (ASV table and representative sequences) were parsed and used as input for PICRUSt2 [31] to generate a table of inferred per-sample abundances of KEGG orthologs. The omixerRpm package [32] (version 0.3.3) was used to reconstruct microbial functionality by computing the abundance of predefined gut metabolic modules (GMM) [33] based on the KO abundances obtained. Differential abundance analysis was performed on GMM abundance tables, following the same approach used for taxonomic counts, to assess cross-sectional and longitudinal associations with UPF consumption.

In relation to fecal metabolomics, rank-based inverse normal transformation (INT) [34] was applied to the data prior to analysis to account for the typically skewed distribution of metabolites and to ensure robustness in subsequent analyses. This method has been previously used in metabolomic studies [35] and performs well with other nonnormal data [36]. The INT involved a two-step procedure: first, the observations were transformed into the probability scale using the empirical cumulative distribution function; second, the observations were transformed into z-scores on the real line.

We used a penalized regression model with the Elastic Net regularization technique [37] as the primary approach to identify metabolites most strongly associated with UPF consumption. This method applies regularization to select only the most stable and reproducible features across multiple iterations. Specifically, we assessed the cross-sectional relationship between UPF consumption and fecal metabolite concentrations at baseline, as well as the longitudinal association between 1-year change in UPF consumption and 1-year changes in fecal metabolite concentrations.

Training and validation procedures were performed. Model training and hyperparameter tuning for alpha and lambda were executed using the caret R package [38]. A resampling strategy of 10-fold cross-validation (10-fold CV) was implemented, with the process repeated 10 times to enhance model reliability. This resampling was conducted on 90% of the dataset designated as the training set, with the remaining 10% held out for final model validation. The best model accuracy was achieved with alpha and lambda values of 0.6 and 36.5, respectively, for the cross-sectional assessment, and with alpha and lambda values of 1 and 34.6, respectively, for the longitudinal assessment. In a subsequent robustness check, 10-fold cross-validation Elastic Net models were run using the defined alpha and lambda values with different random seeds. Only metabolites with coefficients consistently different from zero across all 10 runs were retained, ensuring the selection of stable predictors for the outcome of interest.

To examine the directionality and effect sizes of these pre-selected metabolites, we conducted linear regression analyses adjusting for covariates. Since the metabolites were already pre-selected based on their association with UPF consumption, applying additional FDR correction at this stage would not meaningfully improve statistical rigor and could introduce artificial bias. Instead, we relied on the prior selection process of Elastic Net to ensure robust feature selection.

Linear regression models were fitted using the same approach as that employed to explore the association between alpha diversity and exposure. Results with P < 0.05 were reported as statistically significant.

Results

General characteristics of the study population

The general baseline characteristics of the study population are presented in Table 1.

Table 1 Baseline characteristics of the study population
Full size table

The baseline and 1-year changes in selected covariates are summarized in Table 2. Participants had a significant decrease in BMI (–1.00 ± 5.09 kg/m2) and waist circumference (–3.17 ± 14.51 cm) after 1-year follow-up. In addition, there was a significant improvement in physical activity (106.70 ± 477.35 METs min/day). After one year, the fiber intake and the NOVA Group 1 consumption significantly increased (3.16 ± 10.92 and 107.06 ± 484.81 g/day, respectively), while the NOVA Group 3 and UPF consumption significantly decreased (–71.55 ± 300.68 and − 61.43 ± 172.51 g/day, respectively).

Table 2 Baseline and 1-year changes in selected anthropometric, physical activity and dietary variables
Full size table

Results of fecal microbiota analysis

We did not observe any significant cross-sectional associations between calculated alpha diversity indices and UPF consumption at baseline (Supplementary Tables 1, 2, 3), nor any longitudinal association between 1-year changes in calculated alpha diversity indices and 1-year change in UPF consumption (Supplementary Tables 4, 5, 6). We did not observe significant variance in gut microbiota composition explained by UPF consumption at baseline (Supplementary Table 7), neither explained by 1-year change in UPF consumption (Supplementary Table 8). Differential abundance analysis revealed a negative cross-sectional association between the abundance of Ruminococcaceae incertae sedis and the UPF consumption (β = − 0.275, adjusted P = 0.047) (Fig. 1). The longitudinal analysis showed a significant decrease in the abundance of Parabacteroides spp. associated with increased UPF consumption after 1-year follow-up (β = − 0.278, adjusted P = 0.002) (Fig. 2).

Fig. 1
figure 1

Differentially abundant taxa associated with baseline ultra-processed foods (UPF) consumption. Multivariable association tested with generalized liner model adjusted for sex, age, education (primary, secondary, tertiary), recruiting center (Alicante, Barcelona, Reus, Valencia), smoking status (never, former, smoker), diabetes, hypertension, hypercholesterolemia prevalence, body mass index, waist circumference, physical activity, alcohol intake, fiber intake, NOVA Group 1, Group 2, and Group 3 foods consumption. Values in x axe indicate UPF consumption in g/day, values in y axe indicate genera centered log-ratio relative abundance with Benjamini-Hochberg adjusted P < 0.05

Full size image
Fig. 2
figure 2

Differentially abundant taxa 1-year change associated with ultra-processed foods (UPF) consumption 1-year change. Multivariable longitudinal association tested with generalized liner model adjusted for time (baseline, 1 year), sex, age, education (primary, secondary, tertiary), recruiting center (Alicante, Barcelona, Reus, Valencia), smoking status (never, former, smoker), diabetes, hypertension, hypercholesterolemia prevalence, body mass index, waist circumference, physical activity, alcohol intake, fiber intake, NOVA Group 1, Group 2, and Group 3 foods consumption. Participants’ ID was specified as random effect. Values in x axe indicate UPF consumption 1-year change in g/day, values in y axe indicate genera centered log-ratio relative abundance with Benjamini-Hochberg adjusted P < 0.05

Full size image

Differential abundance analysis of GMM did not yield significant results, either cross-sectionally or longitudinally.

Results of fecal metabolomics analysis

A total of 21 metabolites were selected across all 10 iterations of the binomial elastic net regression cross-validation for the baseline UPF consumption (Fig. 3). Specifically, 11 metabolites presented negative coefficients, and 10 metabolites presented positive coefficients. The linear regression analysis adjusted for covariates returned two metabolites significantly associated with UPF consumption (Supplementary Table 9). Specifically, we observed a negative association with propionylcarnitine (β = − 0.0003, P = 0.013) and L-Pipecolic acid (β = − 0.0003, P = 0.040) (Fig. 4).

Fig. 3
figure 3

Mean coefficients of baseline fecal metabolites concentration selected 10 times in the 10-fold cross-validations of the binomial elastic net regression for the baseline ultra-processed foods consumption

Full size image
Fig. 4
figure 4

Baseline fecal metabolites selected by the binomial elastic net regression significantly associated with baseline ultra-processed foods (UPF) consumption. Association tested with linear regression adjusted for sex, age, education (primary, secondary, tertiary), recruiting center (Alicante, Barcelona, Reus, Valencia), smoking status (never, former, smoker), diabetes, hypertension, hypercholesterolemia prevalence, body mass index, waist circumference, physical activity, alcohol intake, fiber intake, NOVA Group 1, Group 2, and Group 3 foods consumption. Values in x axe indicate UPF consumption in g/day, values in y axe indicate rank-based inverse normal transformation of metabolite concentration with P < 0.05

Full size image

Six metabolites were selected when analyzing the 1-year change in UPF consumption using binomial elastic net regression across all 10 iterations of the cross-validation. (Fig. 5). Specifically, four metabolites presented negative coefficients, and two metabolites presented positive coefficients. However, the linear regression analysis did not show any statistically significant association between these metabolites and the 1-year change in UPF consumption (Supplementary Table 10).

Fig. 5
figure 5

Mean coefficients of fecal metabolites concentration 1-year change selected 10 times in the 10-fold cross-validations of the binomial elastic net regression for the ultra-processed foods consumption 1-year change

Full size image

Discussion

Our study explored the link between UPF consumption, the fecal microbiota composition and the fecal metabolites concentrations in a population of Mediterranean older adults at high cardiovascular risk, within the framework of the PREDIMED-Plus trial. We found that higher UPF consumption was associated with alterations in both fecal microbiota and metabolites, suggesting additional insights into how UPF consumption may impair metabolic health.

At baseline, higher UPF consumption was associated with lower abundance of Ruminococcaceae incertae sedis. While some members of the Ruminococcaceae family are known to contribute to short-chain fatty acid (SCFA) production, the role of Ruminococcaceae incertae sedis remains unclear. Therefore, the observed reduction suggests a potential link to reduced short-chain fatty acids (SCFA) production [39], although this interpretation remains speculative and requires further investigation. This finding aligns with previous research indicating that UPF are typically low in fiber [40], a key substrate for SCFA production [40, 41]. However, the relationship between UPF consumption and Ruminococcaceae may involve mechanisms beyond fiber intake alone. Reduced SCFA production has been shown to negatively affect host energy metabolism and immune function [42, 43], as SCFA have anti-inflammatory properties and serve as an energy source for colonic epithelial cells [44].

Additionally, higher UPF consumption was associated with a lower fecal concentration of propionylcarnitine. This could potentially be explained by reduced fiber intake and the consequent decrease in microbial propionate production [45]. Propionyl-CoA, a precursor of propionylcarnitine, is an intermediate in microbial propionate production [46, 47]. The propanediol pathway relies on specific gut microbes, including members of the Ruminococcaceae family, to convert dietary fiber into propionate [47, 48]. The observed reduction in both fecal Ruminococcaceae incertae sedis abundance and propionylcarnitine concentration suggest that UPF consumption may impair pathways related to SCFA production, potentially leading to lower SCFA availability. Propionylcarnitine also contributes to mitochondrial energy metabolism by maintaining the mitochondrial acyl-CoA/CoA ratio and potentially stimulating the tricarboxylic acid cycle [49]. It indirectly supports mitochondrial function and energy production [49]. Reduced fecal concentrations of propionylcarnitine may indicate impaired mitochondrial function and altered energy metabolism due to UPF consumption [50], though further mechanistic studies are needed to confirm these mechanisms.

We also observed that higher UPF consumption was linked with lower fecal concentrations of pipecolic acid, a non-proteinogenic amino acid involved in lysine catabolism [51]. Pipecolic acid is partially produced by certain gut bacteria through lysine breakdown, and a disrupted microbiota could result to reduced production of this metabolite [52]. UPF are often characterized by lower overall nutritional quality, including reduced protein content [53, 54], which could potentially limit lysine availability. However, the relationship between UPF consumption and pipecolic acid levels may not be fully explained by protein intake alone, as other factors, such as inflammation or gut barrier dysfunction, could also play a role [55,56,57], potentially affecting the conversion of lysine to pipecolic acid. Given its role in immune modulation, reduced pipecolic acid levels may contribute to inflammatory processes [58].

Our findings further suggest that UPF consumption may disrupt amino acid metabolism, as indicated by the observed reduction in pipecolic acid. This disruption could be linked to negative effect of UPF on mitochondrial function, given the central role of mitochondria in amino acid metabolism [50]. UPF often contain advanced glycation end-products (AGE), which can impair mitochondrial function and lead to oxidative stress [50, 59]. These AGE, along with other food additives found in UPF, may exacerbate gut dysbiosis and inflammation, particularly through the accumulation of reactive metabolites that can damage cellular structures, including mitochondria [59]. The observed reduction in pipecolic acid could be indicative of this mitochondrial dysfunction, as mitochondria are involved in its production. However, the direct relationship between UPF consumption and fecal pipecolic acid concentrations requires further investigation. Future studies should explore the links between UPF consumption, lysine intake, gut microbiome alterations, and pipecolic acid production and excretion.

Over one year, increased UPF consumption was associated with reduced abundance of Parabacteroides spp., a genus linked with beneficial metabolic effects, dietary fiber intake, and SCFA production [60, 61]. Our findings suggest that long-term UPF consumption may promote gut dysbiosis, potentially worsening metabolic health. However, longitudinal research on specific microbial genera remains limited, and longer follow-up periods are needed to confirm these findings.

Binomial elastic net regression identified six metabolites associated with changes in UPF consumption over one year. Four metabolites (allopurinol riboside, sphingomyelin, xanthine, 7-dehidrodesmosterol) showed negative coefficients, while two metabolites (glyceric acid, tartaric acid) showed positive coefficients, indicating distinct metabolic responses. However, linear regression analyses did not confirm significant relationships, highlighting the need for cautious interpretation.

Our study benefits from a well-characterized cohort with diverse Spain geographic representation, enhancing the generalizability of our findings within large at-risk populations. The integration of 16 S rRNA sequencing and metabolomics provides a comprehensive multi-dimensional view of both microbial and metabolic alterations linked to UPF consumption. Furthermore, the longitudinal design provides a unique opportunity to examine temporal associations, enhancing our understanding of how UPF consumption might influence gut microbiota composition and metabolic outcomes over time.

The NOVA classification provides a widely used framework for assessing food processing levels, however, it does not differentiate between nutritionally distinct UPFs, such as fiber-rich vs. nutrient-poor products. This heterogeneity may complicate the interpretation of the findings. Our statistical models were adjusted for fiber intake to account for this potential confounding, and the consistency of our findings suggests that fiber content alone does not fully explain the observed associations. Future studies should subclassify UPFs based on their nutritional quality to better understand their impact on gut health and metabolism.

The observational nature of the study limits causal inference, and the lack of longitudinal associations consistent with the cross-sectional findings underscores the need for caution when interpreting these results. Future randomized dietary interventions are needed to confirm the observed associations and clarify the temporal relationships between UPF consumption and gut health.

The inclusion of numerous covariates, while justified by their theoretical relevance, may introduce bias due to collinearity or unmeasured confounders. Although some covariates did not significantly contribute to microbiota variability in PERMANOVA analyses, they were retained to ensure consistency and control for potential confounding. Additional limitations include the limited generalizability of the findings to other populations and the use of 16 S rRNA sequencing, which provides only genus-level taxonomic profiling. Future research should employ multi-omics approaches in more diverse populations.

The results related to the fecal metabolome profile should be interpreted with caution, as gut metabolites can arise from the interaction between dietary metabolites, secondary metabolites produced by gut microbiota, and endogenous metabolites secreted by enterocytes and associated immune cells.

Finally, FFQ may misclassify UPF intake due to self-report bias, leading to potential over- or underestimations across NOVA categories. Future studies should incorporate alternative dietary assessment tools (e.g., 24-hour recalls, food diaries) for improved accuracy.

Conclusions

Our study highlights significant associations between UPF consumption and alterations in fecal microbiota composition and metabolite concentrations, providing valuable insights into the impact of UPF consumption on gut health.

While our study is observational, it suggests that UPF consumption may contribute to metabolic dysregulation and inflammation, particularly in vulnerable populations, warranting further investigation into the underlying mechanisms. Future analyses should explore the potential relationships between these microbial and metabolic alterations to further elucidate the link between diet, gut health, and cardiometabolic outcomes.

From a dietary perspective, our findings suggest that interventions targeting the gut microbiota, such as increasing fiber intake or following a Mediterranean-style diet, may help mitigate the negative metabolic and inflammatory effects of UPF consumption. Choosing whole, minimally processed foods rich in fiber and essential nutrients could potentially counteract some of the negative effects associated with UPFs.

Data availability

The datasets generated and analysed during the current study are not publicly available due to data regulations and for ethical reasons, considering that this information might compromise research participants’ consent because our participants only gave their consent for the use of their data by the original team of investigators. However, collaboration for data analyses can be requested by sending a letter to the PREDIMED-Plus steering Committee (predimed_plus_scom-mittee@googlegroups.com). The request will then be passed to all the members of the PREDIMED-Plus Steering Committee for deliberation.

Abbreviations

UPF:

Ultra-processed foods

MedDiet:

Mediterranean diet

PREDIMED:

PREvención con DIetaMEDiterránea

BMI:

Body mass index

FFQ:

Food frequency questionnaire

ASV:

Amplicon sequence variants

PERMANOVA:

Permutational multivariate analysis of variance

GMM:

Gut metabolic modules

INT:

Inverse normal transformation

SCFA:

Short-chain fatty acid

AGE:

Advanced glycation end-product

References

  1. Monteiro CA, Cannon G, Levy R, Moubarac JC, Jaime P, Martins AP et al. NOVA. The Star Shines Bright (Food Classification. Public Health). World Nutrition. 2016;7(1–3):28–38. Available from: https://worldnutritionjournal.org/index.php/wn/article/view/5

  2. Monteiro CA, Cannon G, Levy RB, Moubarac JC, Louzada ML, Rauber F et al. Ultra-processed foods: what they are and how to identify them. Public Health Nutr. 2019;22(5):936–41. Available from: https://www.cambridge.org/core/product/identifier/S1368980018003762/type/journal_article

  3. Lane MM, Gamage E, Du S, Ashtree DN, McGuinness AJ, Gauci S et al. Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses. BMJ. 2024;e077310. Available from: https://www.bmj.com/lookup/doi/https://doiorg.publicaciones.saludcastillayleon.es/10.1136/bmj-2023-077310

  4. Dicken SJ, Batterham RL. Ultra-processed Food and Obesity: What Is the Evidence? Curr Nutr Rep. 2024;13(1):23–38. Available from: https://link.springer.com/10.1007/s13668-024-00517-z

  5. Juul F, Vaidean G, Parekh N. Ultra-processed Foods and Cardiovascular Diseases: Potential Mechanisms of Action. Advances in Nutrition. 2021;12(5):1673–80. Available from: https://linkinghub.elsevier.com/retrieve/pii/S2161831322004628

  6. Costa de Miranda R, Rauber F, Levy RB. Impact of ultra-processed food consumption on metabolic health. Curr Opin Lipidol. 2021;32(1):24–37. Available from: https://journals.lww.com/https://doiorg.publicaciones.saludcastillayleon.es/10.1097/MOL.0000000000000728

  7. Zinöcker M, Lindseth I. The Western Diet–Microbiome-Host Interaction and Its Role in Metabolic Disease. Nutrients. 2018;10(3):365. Available from: https://www.mdpi.com/2072-6643/10/3/365

  8. García-Montero C, Fraile-Martínez O, Gómez-Lahoz AM, Pekarek L, Castellanos AJ, Noguerales-Fraguas F et al. Nutritional Components in Western Diet Versus Mediterranean Diet at the Gut Microbiota–Immune System Interplay. Implications for Health and Disease. Nutrients. 2021;13(2):699. Available from: https://www.mdpi.com/2072-6643/13/2/699

  9. Matenchuk BA, Mandhane PJ, Kozyrskyj AL. Sleep, circadian rhythm, and gut microbiota. Sleep Med Rev. 2020;53:101340. Available from: https://linkinghub.elsevier.com/retrieve/pii/S1087079220300836

  10. Clark A, Mach N. Exercise-induced stress behavior, gut-microbiota-brain axis and diet: a systematic review for athletes. J Int Soc Sports Nutr. 2016;13(1). Available from: https://www.tandfonline.com/doi/full/https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12970-016-0155-6

  11. Cuevas-Sierra A, Milagro FI, Aranaz P, Martínez JA, Riezu-Boj JI. Gut microbiota differences according to ultra-processed food consumption in a Spanish population. Nutrients. 2021;13(8):1–20.

    Article  Google Scholar 

  12. Atzeni A, Martínez MÁ, Babio N, Konstanti P, Tinahones FJ, Vioque J et al. Association between ultra-processed food consumption and gut microbiota in senior subjects with overweight/obesity and metabolic syndrome. Front Nutr. 2022;9. Available from: https://www.frontiersin.org/articles/https://doiorg.publicaciones.saludcastillayleon.es/10.3389/fnut.2022.976547/full

  13. Muralidharan J, Moreno-Indias I, Bulló M, Lopez JV, Corella D, Castañer O, et al. Effect on gut microbiota of a 1-y lifestyle intervention with mediterranean diet compared with energy-reduced mediterranean diet and physical activity promotion: PREDIMED-Plus study. Am J Clin Nutr. 2021;114(3):1148–58.

    Article  PubMed  PubMed Central  Google Scholar 

  14. García-Gavilán JF, Atzeni A, Babio N, Liang L, Belzer C, Vioque J, et al. Effect of 1-year lifestyle intervention with energy-reduced mediterranean diet and physical activity promotion on the gut metabolome and microbiota: a randomized clinical trial. Am J Clin Nutr. 2024;119(5):1143–54.

    Article  PubMed  Google Scholar 

  15. Brichacek AL, Florkowski M, Abiona E, Frank KM. Ultra-Processed Foods: A Narrative Review of the Impact on the Human Gut Microbiome and Variations in Classification Methods. Nutrients. 2024;16(11):1738. Available from: https://www.mdpi.com/2072-6643/16/11/1738

  16. Martínez-González MA, Buil-Cosiales P, Corella D, Bulló M, Fitó M, Vioque J, et al.Cohort Profile: Design and methods of the PREDIMED-Plus randomized trial. Martínez-González, M A, Buil-Cosiales, P, Corella, D, Bulló, M, Fitó, M, Vioque, J,… Fandos Sánchez, M (2019) Cohort Profile: Design and methods of the PREDIMED-Plus randomized trial International Journal of Epidemiology, 48(2), 387-388o. 2019;48(2):387–388o.Available from: https://academic.oup.com/ije/article/48/2/387/5202210.

  17. Alberti K, Eckel RH, Grundy SM, Zimmet PZ, Cleeman JI, Donato KA et al. Harmonizing the metabolic syndrome A joint interim statement of the international diabetes federation task force on epidemiology and prevention for the study of obesity. Circulation. 2009;120(16):1640–1645.

  18. Molina L, Sarmiento M, Peñafiel J, Donaire D, Garcia-Aymerich J, Gomez M et al. Validation of the Regicor Short Physical Activity Questionnaire for the Adult Population. Lucía A, editor. PLoS One. 2017;12(1):e0168148. Available from: https://doiorg.publicaciones.saludcastillayleon.es/10.1371/journal.pone.0168148

  19. Fernández-Ballart JD, Piñol JL, Zazpe I, Corella D, Carrasco P, Toledo E et al. Relative validity of a semi-quantitative food-frequency questionnaire in an elderly Mediterranean population of Spain. British Journal of Nutrition. 2010;103(12):1808–16. Available from: https://www.cambridge.org/core/product/identifier/S0007114509993837/type/journal_article

  20. Moreiras O, Carbajal Á, Cabrera L, Cuadrado C. Tablas de composición de alimentos. 19a edició. Pirámide. 2018;496.

  21. de la Fuente-Arrillaga C, Vázquez Ruiz Z, Bes-Rastrollo M, Sampson L, Martinez-González MA. Reproducibility of an FFQ validated in Spain. Public Health Nutr. 2010;13(9):1364–72. Available from: https://www.cambridge.org/core/product/identifier/S1368980009993065/type/journal_article

  22. Callahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJA, Holmes SP. DADA2: High-resolution sample inference from illumina amplicon data. Nat Methods. 2016;13(7):581–583. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/nmeth.3869

  23. Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res. 2012;41(D1):D590–6. Available from: http://academic.oup.com/nar/article/41/D1/D590/1069277/The-SILVA-ribosomal-RNA-gene-database-project

  24. Willett W, Stampfer MJ, Total energy intake:. Implications for epidemiologic analyses. Am J Epidemiol. 1986;124(1):17–27. Available from: https://academic.oup.com/aje/article/150019/TOTAL

  25. Chao A. Estimating the Population Size for Capture-Recapture Data with Unequal Catchability. Biometrics. 1987;43(4):783. Available from: https://www.jstor.org/stable/2531532?origin=crossref

  26. Shannon CE. A Mathematical Theory of Communication. Bell System Technical Journal. 1948;27(4):623–56. Available from: https://ieeexplore.ieee.org/document/6773067

  27. Simpson EH. Measurment of Diversity. Nature. 1949;163(1943):688. Available from: https://doiorg.publicaciones.saludcastillayleon.es/10.1038/163688a0

  28. Aitchison J, Barceló-Vidal C, Martín-Fernández JA, Pawlowsky-Glahn V. Logratio analysis and compositional distance. Math Geol. 2000;32(3):271–5.

    Article  Google Scholar 

  29. Anderson MJ. A new method for non-parametric multivariate analysis of variance. Austral Ecol. 2001;26(1):32–46. Available from: https://onlinelibrary.wiley.com/doi/https://doiorg.publicaciones.saludcastillayleon.es/10.1111/j.1442-9993.2001.01070.pp.x

  30. Mallick H, Rahnavard A, McIver LJ, Ma S, Zhang Y, Nguyen LH et al. Multivariable association discovery in population-scale meta-omics studies. PLoS Comput Biol. 2021;17(11):1–27. Available from: https://doiorg.publicaciones.saludcastillayleon.es/10.1371/journal.pcbi.1009442

  31. Douglas GM, Maffei VJ, Zaneveld JR, Yurgel SN, Brown JR, Taylor CM et al. PICRUSt2 for prediction of metagenome functions. Nat Biotechnol. 2020;38(6):685–8. Available from: http://www.nature.com/articles/s41587-020-0548-6

  32. Darzi Y, Falony G, Vieira-Silva S, Raes J. Towards biome-specific analysis of meta-omics data. ISME J. 2016;10(5):1025–8. Available from: https://academic.oup.com/ismej/article/10/5/1025-1028/7538249

  33. Vieira-Silva S, Falony G, Darzi Y, Lima-Mendez G, Garcia Yunta R, Okuda S et al. Species-function relationships shape ecological properties of the human gut microbiome. Nat Microbiol. 2016;1(8):1–8. Available from: https://doiorg.publicaciones.saludcastillayleon.es/10.1038/nmicrobiol.2016.88

  34. McCaw ZR, Lane JM, Saxena R, Redline S, Lin X. Operating characteristics of the rank-based inverse normal transformation for quantitative trait analysis in genome‐wide association studies. Biometrics. 2020;76(4):1262–72. Available from: https://academic.oup.com/biometrics/article/76/4/1262-1272/7425713

  35. García-Gavilán JF, Babio N, Toledo E, Semnani-Azad Z, Razquin C, Dennis C et al. Olive oil consumption, plasma metabolites, and risk of type 2 diabetes and cardiovascular disease. Cardiovasc Diabetol. 2023;22(1):340. Available from: https://biomedcentral-cardiab.publicaciones.saludcastillayleon.es/articles/https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12933-023-02066-1

  36. Bishara AJ, Hittner JB. Testing the significance of a correlation with nonnormal data: Comparison of Pearson, Spearman, transformation, and resampling approaches. Psychol Methods. 2012;17(3):399–417. Available from: https://doiorg.publicaciones.saludcastillayleon.es/10.1037/a0028087

  37. Zou H, Hastie T, Addendum. Regularization and Variable Selection Via the Elastic Net. J R Stat Soc Series B Stat Methodol. 2005;67(5):768–768. Available from: https://academic.oup.com/jrsssb/article/67/5/768/7109667

  38. Kuhn M. Building Predictive Models in R Using the caret Package. J Stat Softw. 2008;28(5). Available from: http://www.jstatsoft.org/v28/i05/

  39. Kim YJ, Jung DH, Park CS. Important roles of Ruminococcaceae in the human intestine for resistant starch utilization. Food Sci Biotechnol. 2024;33(9):2009–19. Available from: https://link.springer.com/https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s10068-024-01621-0

  40. Bulsiewicz WJ. The Importance of Dietary Fiber for Metabolic Health. Am J Lifestyle Med. 2023;17(5):639–48. Available from: http://journals.sagepub.com/doi/10.1177/15598276231167778

  41. Silva YP, Bernardi A, Frozza RL. The Role of Short-Chain Fatty Acids From Gut Microbiota in Gut-Brain Communication. Front Endocrinol (Lausanne). 2020;11. Available from: https://www.frontiersin.org/article/https://doiorg.publicaciones.saludcastillayleon.es/10.3389/fendo.2020.00025/full

  42. den Besten G, van Eunen K, Groen AK, Venema K, Reijngoud DJ, Bakker BM. The role of short-chain fatty acids in the interplay between diet, gut microbiota, and host energy metabolism. J Lipid Res. 2013;54(9):2325–40. Available from: https://linkinghub.elsevier.com/retrieve/pii/S0022227520351245

  43. Liu X, feng, Shao J hao, Liao YT, Wang LN, Jia Y, Dong P, jun et al. Regulation of short-chain fatty acids in the immune system. Front Immunol. 2023;14. Available from: https://www.frontiersin.org/articles/https://doiorg.publicaciones.saludcastillayleon.es/10.3389/fimmu.2023.1186892/full

  44. Ney LM, Wipplinger M, Grossmann M, Engert N, Wegner VD, Mosig AS. Short chain fatty acids: key regulators of the local and systemic immune response in inflammatory diseases and infections. Open Biol. 2023;13(3). Available from: https://doiorg.publicaciones.saludcastillayleon.es/10.1098/rsob.230014

  45. Fu J, Zheng Y, Gao Y, Xu W. Dietary Fiber Intake and Gut Microbiota in Human Health. Microorganisms. 2022;10(12):2507. Available from: https://www.mdpi.com/2076-2607/10/12/2507

  46. Wanders RJA, Waterham HR. Biochemistry of Mammalian Peroxisomes Revisited. Annu Rev Biochem. 2006;75(1):295–332. Available from: https://www.annualreviews.org/doi/https://doiorg.publicaciones.saludcastillayleon.es/10.1146/annurev.biochem.74.082803.133329

  47. Reichardt N, Duncan SH, Young P, Belenguer A, McWilliam Leitch C, Scott KP et al. Phylogenetic distribution of three pathways for propionate production within the human gut microbiota. ISME J. 2014;8(6):1323–35. Available from: https://academic.oup.com/ismej/article/8/6/1323-1335/7582554

  48. El Hage R, Hernandez-Sanabria E, Calatayud Arroyo M, Props R, Van de Wiele T. Propionate-Producing Consortium Restores Antibiotic-Induced Dysbiosis in a Dynamic in vitro Model of the Human Intestinal Microbial Ecosystem. Front Microbiol. 2019;10. Available from: https://www.frontiersin.org/article/https://doiorg.publicaciones.saludcastillayleon.es/10.3389/fmicb.2019.01206/full

  49. Siliprandi N, Lisa F, Di, Menabò R. Propionyl-L-carnitine: Biochemical significance and possible role in cardiac metabolism. Cardiovasc Drugs Ther. 1991;5(1):11–5. Available from: http://link.springer.com/https://doiorg.publicaciones.saludcastillayleon.es/10.1007/BF00128238

  50. Coppola S, Paparo L, Trinchese G, Rivieri AM, Masino A, De Santa Severina G et al. AF,. Increased dietary intake of ultraprocessed foods and mitochondrial metabolism alterations in pediatric obesity. Sci Rep. 2023;13(1):12609. Available from: https://www.nature.com/articles/s41598-023-39566-9

  51. Neshich IAP, Kiyota E, Arruda P. Genome-wide analysis of lysine catabolism in bacteria reveals new connections with osmotic stress resistance. ISME J. 2013;7(12):2400–10. Available from: https://academic.oup.com/ismej/article/7/12/2400/7590259

  52. Li H, Xiao H, yun, Yuan L ping, Yan B, Pan Y, Tian P, ping et al. Protective effect of L-pipecolic acid on constipation in C57BL/6 mice based on gut microbiome and serum metabolomic. BMC Microbiol. 2023;23(1):144. Available from: https://biomedcentral-bmcmicrobiol.publicaciones.saludcastillayleon.es/articles/https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12866-023-02880-3

  53. Hosseinpour-Niazi S, Niknam M, Amiri P, Mirmiran P, Ainy E, Izadi N et al. The association between ultra-processed food consumption and health-related quality of life differs across lifestyle and socioeconomic strata. BMC Public Health. 2024;24(1):1955. Available from: https://biomedcentral-bmcpublichealth.publicaciones.saludcastillayleon.es/articles/https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12889-024-19351-7

  54. Martínez Steele E, Raubenheimer D, Simpson SJ, Baraldi LG, Monteiro CA. Ultra-processed foods, protein leverage and energy intake in the USA. Public Health Nutr. 2018;21(1):114–24. Available from: https://www.cambridge.org/core/product/identifier/S1368980017001574/type/journal_article

  55. Stolfi C, Pacifico T, Monteleone G, Laudisi F. Impact of Western Diet and Ultra-Processed Food on the Intestinal Mucus Barrier. Biomedicines. 2023;11(7):2015. Available from: https://www.mdpi.com/2227-9059/11/7/2015

  56. Groschwitz KR, Hogan SP. Intestinal barrier function: Molecular regulation and disease pathogenesis. Journal of Allergy and Clinical Immunology. 2009;124(1):3–20. Available from: https://linkinghub.elsevier.com/retrieve/pii/S0091674909008641

  57. Ji Y, Sakata Y, Tso P. Nutrient-induced inflammation in the intestine. Curr Opin Clin Nutr Metab Care. 2011;14(4):315–21. Available from: http://journals.lww.com/00075197-201107000-00002

  58. Zhang N, Wang X, Feng M, Li M, Wang J, Yang H et al. Early-life exercise induces immunometabolic epigenetic modification enhancing anti-inflammatory immunity in middle-aged male mice. Nat Commun. 2024;15(1):3103. Available from: https://www.nature.com/articles/s41467-024-47458-3

  59. Martínez Leo EE, Peñafiel AM, Hernández Escalante VM, Cabrera Araujo ZM. Ultra-processed diet, systemic oxidative stress, and breach of immunologic tolerance. Nutrition. 2021;91–92:111419. Available from: https://linkinghub.elsevier.com/retrieve/pii/S0899900721002811

  60. Sun Y, Nie Q, Zhang S, He H, Zuo S, Chen C et al. Parabacteroides distasonis ameliorates insulin resistance via activation of intestinal GPR109a. Nat Commun. 2023;14(1):7740. Available from: https://www.nature.com/articles/s41467-023-43622-3

  61. Medawar E, Haange SB, Rolle-Kampczyk U, Engelmann B, Dietrich A, Thieleking R et al. Gut microbiota link dietary fiber intake and short-chain fatty acid metabolism with eating behavior. Transl Psychiatry. 2021;11(1):500. Available from: https://www.nature.com/articles/s41398-021-01620-3

Download references

Acknowledgements

The authors thank all PREDIMED-Plus participants and investigators. CIBEROBN, CIBERESP, and CIBERDEM are initiative of the Instituto de Salud Carlos III (ISCIII), Madrid, Spain. The authors also thank the PREDIMED-Plus Biobank Network as a part of the National Biobank Platform of the ISCIII for storing and managing the PREDIMED-Plus biological samples.

Funding

This work was supported by the official Spanish Institutions for funding scientific biomedical research, CIBER Fisiopatología de la Obesidad y Nutrición (CIBEROBN) and Instituto de Salud Carlos III (ISCIII), through the Fondo de Investigación para la Salud (FIS), which is co-funded by the European Regional Development Fund (six coordinated FIS projects leaded by J.S-S and J.Vi, including the following projects: PI13/00233, PI13/00728, PI13/00462, PI14/01206, PI14/ 00696, PI16/00533, PI16/00366, PI16/00501, PI17/01441, PI17/00855, PI19/00017, PI19/00781, PI19/00576, PI20/ 00557, PI21/0046; the Especial Action Project entitled: Implementación y evaluación de una intervención intensiva sobre la actividad física Cohorte PREDIMED-Plus grant to J.S-S; the Recercaixa (number 2013ACUP00194) grant to J.S-S; grants from the Consejería de Salud de la Junta de Andalucía (PI0458/2013, PS0358/2016, PI0137/2018); the PROMETEO/ 2017/017 and PROMETEO/2021/21 grants from the Conselleria de Innovación, Universidades, Ciencia y Sociedad Digital from the Generalitat Valenciana; and by NIH grant R01DK127601. This research was also partially funded by the Eat2beNICE/H2020-SFS-2016-2 EU- H2020 European grant, and the Horizon 2020 PRIME study (Prevention and Remediation of Insulin Multimorbidity in Europe; grant agreement #847879). J.S-S, was partially supported by ICREA under the ICREA Academia program. A.H-C is supported by a predoctoral grant from Martí Franquès – INVESTIGO research fellowship funded and supported by NextGenerationEU, Servicio Público de Empleo Estatal and Universitat Rovira i Virgili (2022PMF-INV-01). I.MI was supported by a Miguel Servet type II grant from Instituto de Salud Carlos III, Madrid, Spain (CPII21/00013). None of the funding sources took part in the design, collection, analysis, interpretation of the data, writing the report, or in the decision to submit the manuscript for publication.

Author information

Authors and Affiliations

  1. Centro de Investigación Biomédica en Red - Fisiopatología de la Obesidad y Nutrición (CIBEROBN), Instituto de Salud Carlos III, Madrid, Spain

    Alessandro Atzeni, Adrián Hernández-Cacho, Nadine Khoury, Nancy Babio, Dolores Corella, Montserrat Fitó, Oscar Coltell, Albert Goday, Isabel Moreno Indias, Miguel Ruiz-Canela, Francisco J. Tinahones & Jordi Salas-Salvadó

  2. Desenvolupament i Salut Mental (ANUT-DSM) Departament de Bioquímica i Biotecnologia, Unitat de Nutrició Humana, Alimentació, Nutrició, Universitat Rovira i Virgili, Reus, Spain

    Alessandro Atzeni, Adrián Hernández-Cacho, Nadine Khoury, Nancy Babio & Jordi Salas-Salvadó

  3. Institut d’Investigació Sanitària Pere Virgili (IISPV), Tarragona, Spain

    Alessandro Atzeni, Adrián Hernández-Cacho, Nadine Khoury, Nancy Babio & Jordi Salas-Salvadó

  4. Laboratory of Microbiology, Wageningen University, Wageningen, Netherlands

    Clara Belzer & Prokopis Konstanti

  5. CIBER de Epidemiología y Salud Pública (CIBERESP), Instituto de Salud Carlos III, Madrid, Spain

    Jesús Vioque & Sandra Gonzales-Palacios

  6. Instituto de Investigación Sanitaria y Biomédica de Alicante, Universidad Miguel Hernández (ISABIAL-UMH), Alicante, Spain

    Jesús Vioque & Sandra Gonzales-Palacios

  7. Department of Preventive Medicine, University of Valencia, Valencia, Spain

    Dolores Corella

  8. Unit of Cardiovascular Risk and Nutrition, Institut Hospital del Mar de Investigaciones Médicas Municipal d’Investigació Médica (IMIM), Barcelona, Spain

    Montserrat Fitó & Albert Goday

  9. Metabolomics Platform, The Broad Institute of MIT and Harvard, Boston, MA, USA

    Clary Clish

  10. CIBER Diabetes y Enfermedades Metabólicas (CIBERDEM), Instituto de Salud Carlos III (ISCIII), Madrid, Spain

    Josep Vidal

  11. Department of Endocrinology, Institut d‘Investigacions Biomédiques August Pi Sunyer (IDIBAPS), Hospital Clinic, University of Barcelona, Barcelona, Spain

    Josep Vidal

  12. Department of Computer Languages and Systems, Universitat Jaume I, Castellón, Spain

    Oscar Coltell

  13. Departament de Medicina, Universitat Autónoma de Barcelona, Barcelona, Spain

    Albert Goday

  14. Department of Endocrinology and Nutrition, Instituto de Investigación Biomédica de Málaga (IBIMA), Hospital Universitario Virgen de la Victoria, Málaga, Spain

    Isabel Moreno Indias & Francisco J. Tinahones

  15. Department of Preventive Medicine and Public Health, University of Navarra, Pamplona, Spain

    Silvia Carlos Chillerón & Miguel Ruiz-Canela

  16. Epidemiología y Salud Pública, Instituto de Investigación Sanitaria de Navarra (IdiSNA), Pamplona, Spain

    Silvia Carlos Chillerón & Miguel Ruiz-Canela

  17. Department of Nutrition, Harvard T. H. Chan School of Public Health, Boston, MA, USA

    Frank B. Hu

  18. Channing Division for Network Medicine, Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA

    Frank B. Hu

Authors
  1. Alessandro Atzeni
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Adrián Hernández-Cacho
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Nadine Khoury
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Nancy Babio
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Clara Belzer
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Jesús Vioque
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Dolores Corella
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Montserrat Fitó
    View author publications

    You can also search for this author inPubMed Google Scholar

  9. Clary Clish
    View author publications

    You can also search for this author inPubMed Google Scholar

  10. Josep Vidal
    View author publications

    You can also search for this author inPubMed Google Scholar

  11. Prokopis Konstanti
    View author publications

    You can also search for this author inPubMed Google Scholar

  12. Sandra Gonzales-Palacios
    View author publications

    You can also search for this author inPubMed Google Scholar

  13. Oscar Coltell
    View author publications

    You can also search for this author inPubMed Google Scholar

  14. Albert Goday
    View author publications

    You can also search for this author inPubMed Google Scholar

  15. Isabel Moreno Indias
    View author publications

    You can also search for this author inPubMed Google Scholar

  16. Silvia Carlos Chillerón
    View author publications

    You can also search for this author inPubMed Google Scholar

  17. Miguel Ruiz-Canela
    View author publications

    You can also search for this author inPubMed Google Scholar

  18. Francisco J. Tinahones
    View author publications

    You can also search for this author inPubMed Google Scholar

  19. Frank B. Hu
    View author publications

    You can also search for this author inPubMed Google Scholar

  20. Jordi Salas-Salvadó
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

All the principal PREDIMED-Plus investigators contributed to the study concept and design and to data extraction from the PREDIMED-Plus participants. A.A, and J.S-S contributed to the concept and design of the present study. A.A drafter the manuscript and performed the statistical analyses with the support of A.H-C and N.K, and the supervision of J.S-S. All authors reviewed the manuscript for important intellectual content and approved the final version to be published.

Corresponding authors

Correspondence to Alessandro Atzeni or Jordi Salas-Salvadó.

Ethics declarations

Ethics approval and consent to participate

This trial was approved by the institutional review board of all participating institutions and was registered on the ISRCTN registry (ISRCTN89898870) on July 24, 2014. All participants provided written informed consent, and the procedures were implemented in accordance with the ethical standards of the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

J.S-S reports serving on the board of and receiving grant support through his institution from the International Nut and Dried Fruit Council, serving on the board of the Instituto Danone Spain and the International Danone institute. None of the other authors declare competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Atzeni, A., Hernández-Cacho, A., Khoury, N. et al. The link between ultra-processed food consumption, fecal microbiota, and metabolomic profiles in older mediterranean adults at high cardiovascular risk. Nutr J 24, 62 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12937-025-01125-5

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12937-025-01125-5

Keywords

Nutrition Journal

ISSN: 1475-2891

Contact us