A short-term, high-caloric diet has prolonged effects on brain insulin action in men

Stephanie Kullmann, Lore Wagner, Robert Hauffe, Anne Kühnel, Leontine Sandforth, Ralf Veit, Corinna Dannecker, Jürgen Machann, Andreas Fritsche, Nobert Stefan, Hubert Preissl, Nils B. Kroemer, Martin Heni, André Kleinridders, Andreas L. Birkenfeld
Nature Metabolism
Institute for Diabetes Research and Metabolic Diseases of the Helmholtz Center Munich at the University of Tübingen, Tübingen, Germany

Table of Contents

Overall Summary

Study Background and Main Findings

This study investigated the impact of a short-term (5-day) high-caloric diet (HCD), consisting of an additional 1,500 kcal/day of ultra-processed snacks, on brain insulin action in healthy-weight men. The research aimed to address a gap in understanding the developmental trajectory of brain insulin responsiveness, a crucial factor in regulating energy metabolism and feeding behavior. The study employed a nonrandomized controlled design with two groups: an HCD group (n=18, 17 completed) and a control group (n=11) maintaining their regular diet. Participants underwent assessments at baseline, immediately after the intervention (follow-up 1), and one week after resuming a regular diet (follow-up 2).

Brain insulin action was assessed using functional magnetic resonance imaging (fMRI) combined with intranasal insulin (INI) administration. The primary finding was a significant increase in brain insulin activity in specific regions (right insular cortex, left rolandic operculum, and right midbrain/pons) in the HCD group compared to the control group at follow-up 1. However, at follow-up 2, the HCD group showed significantly lower brain insulin activity in different regions (right hippocampus and bilateral fusiform gyrus) compared to the control group. These changes occurred without significant changes in body weight or peripheral insulin sensitivity, suggesting that brain insulin resistance can develop rapidly in response to dietary changes, preceding changes in overall body composition. The HCD also reduced reward sensitivity and increased punishment sensitivity.

The study concludes that brain insulin responsiveness can adapt to short-term dietary changes before noticeable weight gain occurs. This adaptation, specifically the development of brain insulin resistance, may contribute to the development of obesity and associated metabolic diseases. The findings highlight the importance of dietary choices on brain health and metabolic regulation, even in the absence of immediate weight changes. The study's scope is limited to healthy-weight male participants, and the non-randomized design is a potential weakness, although mitigated by the inclusion of a control group and repeated measures.

Research Impact and Future Directions

The study provides compelling evidence for a rapid and dynamic impact of a short-term, high-caloric diet on brain insulin action, independent of changes in body weight or peripheral insulin sensitivity. Crucially, while the study demonstrates associations between dietary changes, brain insulin activity, and reward processing, it cannot definitively establish causation. The observed correlations between brain insulin activity, liver fat, and reward learning are suggestive, but further research is needed to confirm a direct causal link. The study is limited in its ability to make causal claims due to its non-randomized design.

The study's findings have significant practical implications, suggesting that even short-term dietary indiscretions, particularly those involving ultra-processed foods, can have rapid and potentially detrimental effects on brain function. This highlights the importance of dietary choices for maintaining not only metabolic health but also brain health, even in individuals who are not overweight or obese. The observed changes in reward and punishment sensitivity suggest that dietary interventions may need to address not only the nutritional content of food but also its impact on reward processing.

While the study provides valuable insights, it's crucial to acknowledge the limitations. The findings are specific to healthy-weight men, and their applicability to women or individuals with different metabolic profiles remains unknown. The study also did not directly measure brain inflammation, a potential mechanism linking the HCD to changes in brain insulin action and white matter integrity. Despite these uncertainties, the study strongly suggests that dietary interventions aimed at preventing or treating obesity and related metabolic disorders should consider the rapid impact of diet on brain function.

Several critical questions remain unanswered. The study does not determine whether the observed effects are due to the excessive calories, the specific macronutrient composition of the diet, or the ultra-processed nature of the snacks. The long-term consequences of these short-term changes in brain insulin action are also unknown. While the nonrandomized design is a significant limitation, the inclusion of a control group and repeated measures strengthens the internal validity. However, the lack of randomization could introduce bias, and it's possible that unmeasured confounding factors contributed to the observed differences between groups. Future studies should address these limitations to confirm and extend the findings.

Critical Analysis and Recommendations

Clear Statement of Main Finding (written-content)
The abstract clearly states the main finding: a short-term HCD disrupts brain insulin action in healthy men, persisting after returning to a regular diet. This concise statement of the core result is important because it immediately informs the reader of the study's primary outcome, enhancing the abstract's impact and readability. This allows readers to quickly grasp the study's significance.
Section: Abstract
Quantify Caloric Increase in HCD (written-content)
The abstract does not quantify the caloric increase in the HCD. Including the specific caloric increase (1,500 kcal/day) would provide a clearer picture of the intervention's intensity, which is crucial for understanding the study's scope and for comparison with other studies. This would allow for better contextualization of the findings.
Section: Abstract
Effective Contextualization of Insulin Resistance (written-content)
The introduction effectively establishes the context by highlighting the detrimental effects of insulin resistance in both the periphery and the central nervous system. This is important because it provides a clear rationale for the study, connecting it to the broader issue of metabolic health and emphasizing the significance of brain insulin action. This sets the stage for the research question.
Section: Introduction
Emphasize Importance of Early Detection/Intervention (written-content)
The introduction does not explicitly state why understanding the early development of brain insulin resistance is crucial. Emphasizing the potential for early intervention and prevention would significantly enhance the introduction's impact by highlighting the clinical and public health relevance of the research. This would strengthen the justification for the study.
Section: Introduction
Increased Brain Insulin Activity After HCD (written-content)
The HCD group exhibited significantly higher brain insulin activity in specific regions (right insular cortex, left rolandic operculum, and right midbrain/pons) immediately after the 5-day diet (follow-up 1) compared to the control group, when adjusted for baseline (P_FWE < 0.05, whole-brain corrected). This finding, obtained using fMRI and intranasal insulin administration, is crucial because it demonstrates a rapid and region-specific effect of the HCD on brain insulin action. This suggests a potential mechanism linking diet to altered brain function, with implications for understanding the development of metabolic disorders.
Section: Results
Specify Statistical Method for Baseline Adjustment (written-content)
The Results section does not explicitly state the statistical method used for baseline adjustment. Specifying the method (e.g., ANCOVA, difference scores) is crucial for methodological transparency and allows readers to fully understand the analysis and assess its appropriateness. This would enhance the rigor and reproducibility of the study.
Section: Results
Detailed Participant Inclusion Criteria (written-content)
The Methods section clearly describes the participant inclusion criteria, including age, BMI, health status, and lifestyle factors. This level of detail is crucial for assessing the study's internal validity (by minimizing confounding factors) and the generalizability of the findings (by defining the specific population studied). This enhances the rigor and interpretability of the study.
Section: Methods
Specify How Step Count Compliance Was Assessed (written-content)
The Methods section does not specify how compliance with the step count restriction (fewer than 4,000 steps/day) was assessed or enforced. Adding this information would strengthen the methods by providing a clearer picture of how physical activity, a potential confounding variable, was controlled. This would improve the study's internal validity.
Section: Methods
Effective Summary of Main Findings (written-content)
The Discussion effectively summarizes the main findings, highlighting the rapid adaptation of brain insulin responsiveness to the HCD and the persistence of these effects. This reinforces the core message of the paper and provides a clear and concise takeaway for the reader. This strengthens the overall impact of the study.
Section: Discussion
Explicitly Discuss Limitations Regarding Inflammation (written-content)
The Discussion could more explicitly discuss the limitations of drawing conclusions about brain inflammation based on the available data (no changes in circulating cytokines immediately after the HCD). Acknowledging this limitation would provide a more balanced perspective and highlight the need for further research with more direct measures of brain inflammation. This would strengthen the critical evaluation of the findings.
Section: Discussion

Section Analysis

Abstract

Key Aspects

Strengths

Suggestions for Improvement

Introduction

Key Aspects

Strengths

Suggestions for Improvement

Non-Text Elements

Table 1 | Participants' metabolic characteristics
Figure/Table Image (Page 2)
Table 1 | Participants' metabolic characteristics
First Reference in Text
Participants completed three visits (baseline, follow-up 1 and follow-up 2) during an assessment period of approximately 3-4 weeks (see Fig. 1 for study design).
Description
  • Overview of metabolic characteristics: Table 1 presents the metabolic characteristics of the participants at baseline and follow-up visits, separating them into control and high-caloric diet (HCD) groups. It includes body composition measures like body weight (approximately 73-74 kg for both groups at baseline), BMI (around 22 kg/m^2), and waist-to-hip ratio (about 0.85-0.86). It also details total and subcutaneous adipose tissue, as well as visceral adipose tissue and liver fat derived from magnetic resonance (MR) techniques. Metabolic parameters include HbA1c (around 5.1-5.2%), Matsuda insulin sensitivity index (around 18), HOMA-IR (around 1.9), fasting insulin (about 54 pmol/L), and glucose levels (around 4.7 mmol/L). Further parameters listed are fasting glucagon, triglycerides, IL-6, CRP, gamma-glutamyl transferase, testosterone, resting energy expenditure (around 2100 kcal), and respiratory quotient (around 0.85).
  • Explanation of key metabolic parameters: HbA1c, or glycated hemoglobin, indicates average blood sugar levels over the past 2-3 months. Matsuda insulin sensitivity index is a measure of how sensitive someone is to insulin, with higher values indicating better insulin sensitivity. HOMA-IR, or Homeostatic Model Assessment for Insulin Resistance, estimates insulin resistance, where lower values are better. IL-6 (Interleukin-6) and CRP (C-reactive protein) are inflammatory markers, and gamma-glutamyl transferase is an enzyme primarily found in the liver.
Scientific Validity
  • Relevance of parameters: The table presents relevant metabolic characteristics that are commonly used to assess metabolic health and insulin sensitivity. The use of MR-derived measures for body fat composition adds rigor to the assessment.
  • Inclusion of baseline and follow-up data: The inclusion of both baseline and follow-up data allows for a comparison of changes within and between the groups. The statistical analysis should appropriately account for within-subject correlations.
  • Adequate sample size: The sample sizes for each group (Control and HCD) are reasonable for detecting differences in metabolic parameters. However, a power analysis should have been conducted to determine the appropriate sample size.
Communication
  • Clear and organized presentation: The table provides a comprehensive overview of the participants' baseline characteristics, which is crucial for understanding the study population. The clear labeling and organization of the table enhance its readability and facilitate comparisons between the control and HCD groups.
  • Accurate caption and inclusion of units: The table's caption accurately reflects its content, aiding readers in quickly understanding the purpose of the table. Including units of measurement for each parameter ensures clarity and avoids ambiguity.
Fig. 1 | Schematic overview of the study design.
Figure/Table Image (Page 3)
Fig. 1 | Schematic overview of the study design.
First Reference in Text
Participants completed three visits (baseline, follow-up 1 and follow-up 2) during an assessment period of approximately 3-4 weeks (see Fig. 1 for study design).
Description
  • Overview of study design: Fig. 1 presents a schematic diagram of the study design, outlining the timeline and procedures for both the control and high-caloric diet (HCD) groups. The study involves three visits: baseline, follow-up 1, and follow-up 2, spanning approximately 3-4 weeks. The HCD group consumes a high-caloric diet for 5 days before follow-up 1, while the control group maintains a regular diet. Key assessments include brain MRI with insulin spray, whole-body MRI, and oral glucose tolerance tests (OGTT). Blood samples are collected, and a reward-learning task is performed. The schematic highlights the timing of each assessment relative to the dietary intervention.
  • Explanation of key procedures: An oral glucose tolerance test (OGTT) is a test to see how well your body processes sugar. It involves drinking a sugary drink and then having your blood sugar levels checked over the next few hours. Brain MRI involves using magnetic fields and radio waves to create detailed images of the brain, while 'insulin spray' indicates intranasal insulin administration.
Scientific Validity
  • Accurate representation of study design: The schematic accurately reflects the study design as described in the text. The inclusion of all relevant procedures and time points enhances the validity of the representation.
  • Appropriate use of schematic diagram: The use of a schematic diagram is appropriate for illustrating the study design, as it provides a clear and concise overview of the experimental timeline and procedures.
Communication
  • Clear visual representation: The figure provides a clear visual representation of the study design, making it easier for readers to understand the experimental timeline and procedures. The use of distinct colors and labels enhances the clarity of the schematic.
  • Effective communication of study sequence: The schematic effectively communicates the sequence of events, including the baseline visits, intervention period (HCD or regular diet), and follow-up visits. The inclusion of key procedures like brain MRI, insulin spray, whole-body MRI, and OGTT helps readers grasp the scope of the study.

Results

Key Aspects

Strengths

Suggestions for Improvement

Non-Text Elements

Fig. 2 | Disrupted brain insulin action after short-term overeating with...
Full Caption

Fig. 2 | Disrupted brain insulin action after short-term overeating with calorie-rich snacks.

Figure/Table Image (Page 4)
Fig. 2 | Disrupted brain insulin action after short-term overeating with calorie-rich snacks.
First Reference in Text
The HCD group had significantly higher insulin activity in parts of the right insular cortex, left rolandic operculum and right midbrain/pons (Fig. 2b) at follow-up 1 adjusted for baseline compared with the control group (PFWE < 0.05, whole-brain corrected, where FWE indicates family wise error; Extended Data Table 4).
Description
  • Overview of brain insulin action data: Fig. 2 presents data on brain insulin action following a high-caloric diet (HCD). Panel A shows brain maps indicating regions with significantly altered insulin activity. Panel B depicts significantly higher insulin activity in the right insular cortex, left rolandic operculum, and right midbrain/pons in the HCD group at follow-up 1 compared to the control group. Panel C illustrates significantly lower insulin activity in the right hippocampus and bilateral fusiform gyrus at follow-up 2. Panel D includes scatter plots showing correlations between changes in brain insulin activity, liver fat, saturated fatty acid (SFA) intake, and reward sensitivity.
  • Explanation of statistical terms: Family Wise Error (FWE) is a statistical correction used to reduce the chance of false positives when conducting multiple comparisons, as is common in brain imaging studies. The scatter plots illustrate correlations, which represent the extent to which two variables tend to change together. A positive correlation means that as one variable increases, the other tends to increase as well, while a negative correlation means that as one variable increases, the other tends to decrease.
Scientific Validity
  • Statistical rigor: The use of FWE correction (PFWE < 0.05) strengthens the validity of the brain imaging results by controlling for multiple comparisons. Adjusting for baseline measurements helps account for individual differences and reduces noise.
  • Correlations and causality: The correlations presented in panel D provide evidence linking brain insulin activity to metabolic and behavioral changes. However, correlation does not imply causation, and further studies are needed to establish causal relationships.
  • Relevance of brain regions: The brain regions identified (insular cortex, rolandic operculum, midbrain/pons, hippocampus, fusiform gyrus) are relevant to insulin signaling, reward processing, and cognitive functions. The observed changes in these regions support the study's hypothesis.
Communication
  • Clear and concise caption: The figure caption clearly indicates the main finding: short-term overeating disrupts brain insulin action. The figure effectively uses color-coded brain maps and scatter plots to present complex data in an accessible manner.
  • Effective visualization of relationships: The figure effectively communicates the relationship between brain insulin activity, liver fat, dietary intake, and reward sensitivity. The use of scatter plots helps visualize correlations, while brain maps highlight specific regions affected by the intervention.
Extended Data Fig. 1 | Liver fat content in the control and high-caloric diet...
Full Caption

Extended Data Fig. 1 | Liver fat content in the control and high-caloric diet group (HCD) at baseline and 5-days after the high-caloric or regular diet intervention (follow-up 1).

Figure/Table Image (Page 10)
Extended Data Fig. 1 | Liver fat content in the control and high-caloric diet group (HCD) at baseline and 5-days after the high-caloric or regular diet intervention (follow-up 1).
First Reference in Text
However, liver fat content increased in the HCD group (group-by-visit interaction, estimate of -0.11, 95% CI -0.19 to -0.03, P = 0.008; HCD group baseline versus HCD follow-up 1, estimate of -0.3744, s.e. = 0.104, d.f. = 30.4, t = 3.6, P = 0.005; Extended Data Fig. 1), whereas it did not change in the control group (P = 0.958).
Description
  • Overview of liver fat content data: Extended Data Fig. 1 shows the liver fat content in the control and high-caloric diet (HCD) groups at baseline and 5 days after the intervention (follow-up 1). Liver fat content increased significantly in the HCD group from baseline to follow-up 1 (p = 0.005). In contrast, liver fat content did not change significantly in the control group (p = 0.958). The data are presented as box plots with whiskers indicating 1.5 times the interquartile range, along with a line diagram showing individual data points.
  • Explanation of box plot: A box plot is a standardized way of displaying the distribution of data based on a five number summary: minimum, first quartile (25th percentile), median, third quartile (75th percentile), and maximum. The whiskers extending from the box indicate the range of the data, excluding outliers. The interquartile range (IQR) is the range between the first and third quartiles and represents the middle 50% of the data.
Scientific Validity
  • Support for study finding: The figure supports the finding that a high-caloric diet leads to a significant increase in liver fat content within a short period. The statistical analysis, as indicated by the p-values, appears appropriate for the data.
  • Appropriate visualization: The use of a box plot and line diagram is suitable for visualizing the distribution of liver fat content and showing individual changes. However, it would be beneficial to include the specific statistical test used to determine the p-values.
  • More details on statistics: Presenting the estimate, confidence interval (CI), standard error (s.e.), degrees of freedom (d.f.), and t-statistic would provide more detailed information regarding the statistical analysis performed and enhance the rigor of the presentation.
Communication
  • Clear visualization of liver fat content: The figure clearly presents the liver fat content data for both the control and HCD groups, allowing for a visual comparison of the changes from baseline to follow-up 1. The use of a box plot with a line diagram enhances the visualization of individual changes within each group.
  • Accurate caption and inclusion of p-values: The figure caption accurately describes the data presented, specifying the groups, time points, and intervention. Including the p-values in the caption provides immediate information about the statistical significance of the observed changes.
Extended Data Fig. 2 | HCD reduced reward sensitivity and increased punishment...
Full Caption

Extended Data Fig. 2 | HCD reduced reward sensitivity and increased punishment sensitivity.

Figure/Table Image (Page 11)
Extended Data Fig. 2 | HCD reduced reward sensitivity and increased punishment sensitivity.
First Reference in Text
Compared with the control group, the HCD reduced reward sensitivity (t(27) = -3.6, Pboot < 0.001, where boot indicates bootstrapping) and increased punishment sensitivity (t(27) = 2.6, Pboot = 0.002) at follow-up 1 (Extended Data Fig. 2).
Description
  • Overview of reward and punishment sensitivity data: Extended Data Fig. 2 shows the effects of a high-caloric diet (HCD) on reward and punishment sensitivity. Compared to the control group, the HCD group exhibited reduced reward sensitivity (t(27) = -3.6, Pboot < 0.001) and increased punishment sensitivity (t(27) = 2.6, Pboot = 0.002) at follow-up 1. The figure presents bootstrapped density plots of the difference between groups in changes of parameter estimates at follow-up 1 and follow-up 2.
  • Explanation of statistical terms: Bootstrapping is a statistical technique used to estimate the sampling distribution of a statistic by resampling from the original data. This allows for the calculation of confidence intervals and p-values without assuming a specific distribution. The 't' value is a measure of the difference between two group means relative to the variability within the groups.
Scientific Validity
  • Statistical rigor: The use of bootstrapping to calculate p-values strengthens the validity of the results, as it does not rely on assumptions about the underlying distribution of the data. Reporting the t-statistic and degrees of freedom provides essential information for interpreting the statistical significance.
  • Support for study finding: The figure supports the finding that a high-caloric diet alters reward and punishment sensitivity. The results are consistent with the hypothesis that overeating can disrupt reward processing.
  • Need for effect sizes: The figure would benefit from including effect sizes (e.g., Cohen's d) to quantify the magnitude of the observed effects. This would provide a more complete picture of the impact of the HCD on reward and punishment sensitivity.
Communication
  • Clear and concise caption: The figure caption clearly states the main finding: that a high-caloric diet (HCD) reduces reward sensitivity and increases punishment sensitivity. This provides a concise summary of the results presented.
  • Effective use of density plots: The figure uses density plots to illustrate the differences in parameter estimates between the HCD and control groups. While density plots are suitable for visualizing distributions, providing additional summary statistics (e.g., means, standard deviations) could enhance clarity.
Extended Data Fig. 3 | Change in white matter organization at follow-up 2.
Figure/Table Image (Page 12)
Extended Data Fig. 3 | Change in white matter organization at follow-up 2.
First Reference in Text
The HCD group had significantly lower FA values mainly located on the inferior fronto-occipital fasciculus, genu of the corpus callosum and anterior corona radiata (Extended Data Fig. 3; P<0.05; threshold-free cluster enhancement (TFCE)-corrected) as well as higher MD values in the superior corona radiata at follow-up 2 compared with baseline.
Description
  • Overview of white matter organization data: Extended Data Fig. 3 shows changes in white matter organization at follow-up 2 in the high-caloric diet (HCD) group compared to the control group. The figure displays the fractional anisotropy (FA) skeleton in green, representing major white matter tracts. Areas with significantly lower FA in the HCD group are shown in red, orange, and yellow (p-corr < 0.05). These areas are mainly located in the inferior fronto-occipital fasciculus, genu of the corpus callosum, and anterior corona radiata. The colorbar represents the 1-p value (TFCE-corrected).
  • Explanation of key terms: Fractional anisotropy (FA) is a measure of the directionality of water diffusion in white matter, reflecting the integrity of the myelin sheath surrounding nerve fibers. Lower FA values indicate reduced white matter integrity. Threshold-Free Cluster Enhancement (TFCE) is a statistical method used to correct for multiple comparisons in imaging data.
Scientific Validity
  • Statistical rigor: The use of TFCE correction strengthens the validity of the results by controlling for multiple comparisons. The figure supports the finding that a high-caloric diet is associated with reduced white matter integrity in specific brain regions.
  • Relevance of white matter tracts: The identified white matter tracts (inferior fronto-occipital fasciculus, genu of the corpus callosum, and anterior corona radiata) are relevant to cognitive and executive functions. The observed changes in these tracts are consistent with the study's hypothesis.
  • Need for MD values: It would be helpful to include information about the mean diffusivity (MD) values in the figure, as the reference text mentions higher MD values in the superior corona radiata. This would provide a more complete picture of the changes in white matter organization.
Communication
  • Clear caption and color-coding: The figure caption clearly indicates that the image depicts changes in white matter organization at follow-up 2, providing context for the presented data. The use of color-coding (red, orange, and yellow) to represent areas with lower fractional anisotropy (FA) enhances visual interpretation.
  • Effective highlighting of white matter tracts: The figure effectively highlights specific white matter tracts, such as the inferior fronto-occipital fasciculus, genu of the corpus callosum, and anterior corona radiata, where significant changes in FA were observed. Annotations on the brain image aid in identifying these structures.
Extended Data Table 1 | Food Diary
Figure/Table Image (Page 13)
Extended Data Table 1 | Food Diary
First Reference in Text
group increased their daily total caloric intake on average by 1,200 kcal between the baseline and follow-up 1 visit (P<0.05; Extended Data Table 1).
Description
  • Overview of food diary data: Extended Data Table 1 presents a detailed food diary analysis, reporting energy intake (in kcal), fat (in % and g), carbohydrates (in % and g), proteins (in % and g), saturated fatty acids (in g), monounsaturated fatty acids (in g), polyunsaturated fatty acids (in g), and fiber (in g) for the control and high-caloric diet (HCD) groups at baseline, follow-up 1, and follow-up 2. The table indicates that the HCD group increased their daily total caloric intake on average by 1,200 kcal between the baseline and follow-up 1 visit (P<0.05). The table also mentions that there were significant within-group differences for the HCD group from baseline to follow-up 1, based on linear mixed effects models (LMER).
  • Explanation of key terms: A kilocalorie (kcal) is a unit of energy, often used to measure the energy content of food. Linear mixed effects models (LMER) are statistical models that incorporate both fixed effects (factors that are directly manipulated or observed) and random effects (factors that vary randomly). They are useful for analyzing data with hierarchical or clustered structures, such as repeated measurements within individuals.
Scientific Validity
  • Value of food diary data: The inclusion of a food diary is valuable for assessing dietary intake and adherence to the intervention. The level of detail provided (macronutrient composition, fiber) enhances the rigor of the analysis.
  • Appropriate statistical analysis: The use of LMER is appropriate for analyzing the repeated measures data. However, it would be beneficial to report the specific LMER model used and the assumptions tested.
  • Omission of dietary assessment method: The table should include a clear statement about the method used for dietary assessment (e.g., weighed food records, 24-hour recalls) and the validity/reliability of the method. The table lacks information on the method used to assess dietary intake, impacting the assessment of validity.
Communication
  • Facilitates comparison but can be overwhelming: The table's structure, presenting data for both control and HCD groups across baseline and follow-up visits, facilitates direct comparison of dietary intake. However, the sheer volume of data might overwhelm some readers; highlighting key differences more prominently could improve clarity.
  • Clear notation of significant differences: The use of asterisks to denote significant within-group differences is helpful, but a more detailed explanation of the statistical tests used would improve transparency.
Extended Data Table 3 | Inflammatory markers
Figure/Table Image (Page 15)
Extended Data Table 3 | Inflammatory markers
First Reference in Text
No significant differences were identified for metabolic parameters (P > 0.05; Table 1), including peripheral insulin sensitivity based on the oral glucose tolerance test (OGTT)-derived Matsuda Index and the homeostasis model assessment insulin resistance (HOMA-IR), as well as inflammatory markers such as C-reactive protein (CRP), interleukin (IL)-6 (Table 1) and other cytokines (Extended Data Table 3).
Description
  • Overview of inflammatory marker data: Extended Data Table 3 presents data on inflammatory markers for the control and high-caloric diet (HCD) groups at baseline and follow-up 1. The table includes VCAM, ICAM, Osteopontin, sTNF-R1, sTNF-R2, Hu IP-10, Hu MCP-1, Hu MIP-1b, Hu RANTES, Hu IFN-g, Hu-TNF-a, Hu-IL-18 and Hu MIG. The table indicates that no significant differences were identified for these inflammatory markers (P > 0.05). The values are time-point normalized and presented in arbitrary units.
  • Explanation of inflammatory markers: VCAM and ICAM are cell adhesion molecules involved in inflammation. sTNF-R1 and sTNF-R2 are soluble TNF receptors. Hu IP-10, Hu MCP-1, Hu MIP-1b, and Hu RANTES are chemokines involved in immune cell recruitment. Hu IFN-g and Hu-TNF-a are cytokines involved in inflammatory responses. Hu-IL-18 and Hu MIG are interleukins.
Scientific Validity
  • Comprehensive assessment: The inclusion of multiple inflammatory markers provides a comprehensive assessment of the inflammatory response to the high-caloric diet. The use of time-point normalized values is appropriate for comparing changes within and between groups.
  • Support for study finding: The table supports the finding that a short-term high-caloric diet does not significantly alter systemic inflammatory markers, despite changes in other metabolic parameters and brain insulin action.
  • Consideration of power: It is important to note that the absence of significant differences does not necessarily mean that there were no effects on inflammation. The study may have lacked the power to detect small changes in inflammatory markers. Providing information on the power of the study to detect changes in these inflammatory markers would enhance the rigor of the presentation.
Communication
  • Clear labeling: The table is clearly labeled, indicating that it contains data on inflammatory markers. However, the lack of significant differences may lead some readers to overlook this table, even though it provides important null findings.
  • Need for emphasis in main text: A brief statement in the main text emphasizing that no significant changes in inflammatory markers were observed, despite changes in other metabolic parameters, would enhance the impact of this table.
Extended Data Table 4 | Changes in brain insulin activity (CBF) from before to...
Full Caption

Extended Data Table 4 | Changes in brain insulin activity (CBF) from before to after the intervention

Figure/Table Image (Page 16)
Extended Data Table 4 | Changes in brain insulin activity (CBF) from before to after the intervention
First Reference in Text
The HCD group had significantly higher insulin activity in parts of the right insular cortex, left rolandic operculum and right midbrain/pons (Fig. 2b) at follow-up 1 adjusted for baseline compared with the control group (PFWE < 0.05, whole-brain corrected, where FWE indicates family wise error; Extended Data Table 4).
Description
  • Overview of brain insulin activity data: Extended Data Table 4 presents changes in brain insulin activity, as measured by cerebral blood flow (CBF), from before to after the intervention. It reports brain regions showing significant differences between the high-caloric diet (HCD) and control groups at follow-up 1 and follow-up 2. For each region, the table lists the hemisphere (Hemi), MNI coordinates (x, y, z), peak t-value, cluster size (k), and FWE-corrected p-value (PFWE). At follow-up 1, the HCD group showed significantly higher activity in the right region in pons/midbrain, right insula, and left rolandic operculum. At follow-up 2, the HCD group showed significantly lower activity in the right hippocampus and bilateral fusiform gyrus.
  • Explanation of neuroimaging terms: MNI coordinates are a standardized coordinate system used in neuroimaging to identify the location of brain structures. The t-value is a measure of the difference between two group means relative to the variability within the groups. The cluster size (k) indicates the number of contiguous voxels showing significant activation. PFWE (family-wise error) is a correction for multiple comparisons.
Scientific Validity
  • Statistical rigor: The table provides essential information for interpreting the brain imaging results. The use of FWE correction strengthens the validity of the findings by controlling for multiple comparisons.
  • Relevance of brain regions: The reported brain regions are relevant to insulin signaling, reward processing, and cognitive functions. The observed changes in these regions support the study's hypothesis.
  • Need for contrast information: It would be helpful to include information about the specific contrasts used to generate these results (e.g., HCD > Control, Control > HCD). This would clarify the direction of the observed effects.
Communication
  • Detailed information on brain regions: The table provides detailed information on the brain regions showing significant changes in cerebral blood flow (CBF) following the intervention. The inclusion of MNI coordinates, peak t-values, and cluster sizes allows for a precise localization and characterization of the observed effects.
  • Effective summarization of findings: The table effectively summarizes the key findings related to brain insulin activity. However, providing a brief explanation of the MNI coordinate system would enhance accessibility for readers unfamiliar with neuroimaging.
Extended Data Table 5 | Correlation between change in brain insulin...
Full Caption

Extended Data Table 5 | Correlation between change in brain insulin responsiveness and metabolic and behavioural changes

Figure/Table Image (Page 17)
Extended Data Table 5 | Correlation between change in brain insulin responsiveness and metabolic and behavioural changes
First Reference in Text
Correlation analyses showed that higher insulin activity at follow-up1was significantly associated with fold change in liver fat, the change in reward learning and the food diary reported fold change in fat and saturated fatty acid intake, especially in the pons/midbrain (Fig. 2d) (P<0.05; Extended Data Table 5).
Description
  • Overview of correlation data: Extended Data Table 5 presents the results of correlation analyses between changes in brain insulin responsiveness and metabolic and behavioral changes. The table reports Pearson correlation coefficients (r) and p-values (p) for associations between insulin activity in specific brain regions (insula, rolandic operculum, pons/midbrain, fusiform gyrus, hippocampus) and variables such as liver fat, reward sensitivity, punishment sensitivity, and dietary intake. The analyses are performed for follow-up 1 and follow-up 2, adjusted for baseline.
  • Explanation of correlation terms: A Pearson correlation coefficient (r) measures the strength and direction of a linear relationship between two variables. It ranges from -1 to +1, where -1 indicates a perfect negative correlation, +1 indicates a perfect positive correlation, and 0 indicates no correlation. A p-value indicates the probability of observing the results if there is no true association between the variables.
Scientific Validity
  • Appropriate statistical method: The use of correlation analyses is appropriate for exploring relationships between variables. However, it is important to note that correlation does not imply causation. Further studies are needed to establish causal relationships.
  • Support for study findings: The table supports the finding that brain insulin responsiveness is associated with metabolic and behavioral changes. The specific correlations identified provide insights into the potential mechanisms linking brain insulin action to liver fat, reward processing, and dietary intake.
  • Need for multiple comparisons correction: The use of a significance threshold of P < 0.05 without correction for multiple comparisons is a limitation. Given the number of correlations tested, there is an increased risk of false positive findings. Applying a correction for multiple comparisons (e.g., Bonferroni correction) would enhance the rigor of the analysis.
Communication
  • Concise summary of correlations: The table provides a concise summary of the correlation analyses, making it easy to identify significant associations between brain insulin responsiveness and other variables. However, the limited information (r and p-value only) might hinder a deeper understanding of the relationships.
  • Need for effect sizes or confidence intervals: The table would benefit from including effect sizes or confidence intervals for the correlations, which would provide a more complete picture of the strength and precision of the associations.

Methods

Key Aspects

Strengths

Suggestions for Improvement

Non-Text Elements

Extended Data Table 2 | State Questionnaires on Brain MRI day in the fasted...
Full Caption

Extended Data Table 2 | State Questionnaires on Brain MRI day in the fasted state

Figure/Table Image (Page 14)
Extended Data Table 2 | State Questionnaires on Brain MRI day in the fasted state
First Reference in Text
Not explicitly referenced in main text
Description
  • Overview of state questionnaire data: Extended Data Table 2 presents results from state questionnaires administered on the brain MRI day in the fasted state. The questionnaires include Desire to eat (VAS), FCQS total scale, PANAS positive affect, and PANAS negative affect. The table reports means and standard deviations for the control and HCD groups at baseline, follow-up 1, and follow-up 2.
  • Explanation of questionnaires: VAS (visual analogue scale) is a measurement instrument that attempts to measure a characteristic or attitude that is believed to range across a continuum of values and cannot easily be directly measured. FCQS (food craving questionnaire state) measures current food cravings. PANAS (positive and negative affect schedule) measures positive and negative emotions and mood.
Scientific Validity
  • Reliable questionnaires but unclear relevance: The use of standardized questionnaires (VAS, FCQS, PANAS) provides a reliable and validated method for assessing subjective states. However, the lack of explicit referencing in the main text raises questions about the relevance and interpretation of these data.
  • Need for clear rationale: Without knowing the specific hypotheses related to these questionnaires, it is difficult to assess the appropriateness of their inclusion. A clear rationale for including these measures and how they relate to the study's objectives is needed.
  • Lack of statistical analyses: The table reports means and standard deviations, but it does not include any statistical analyses comparing the groups or visits. Without statistical analyses, it is difficult to draw any firm conclusions from these data.
Communication
  • Potential value but unclear contribution: The table presents data from state questionnaires, which can be valuable for understanding the participants' subjective experiences during the study. However, the lack of explicit referencing in the main text makes it difficult to assess the table's direct contribution to the study's key findings.
  • Need for more descriptive title and explanation: The table could benefit from a more descriptive title, such as "State Questionnaire Results on Brain MRI Day," to enhance clarity. Including a brief explanation of each questionnaire (e.g., FCQS, PANAS, VAS) in the caption or a footnote would also improve understanding.

Discussion

Key Aspects

Strengths

Suggestions for Improvement

↑ Back to Top