The pentose phosphate pathway (PPP) furnishes NADPH for hepatic de novo lipogenesis, driving intrahepatic lipid (IHL) accumulation. The aim of the current study was to examine the association between the PPP, proxied by 24-h urinary erythritol, and IHL content at the population level. We used cross-sectional data from the Maastricht Study, a population-based cohort study (N = 1,494; mean ± SD age 59 ± 8 years; 49% women). We first assessed the relationship between 24-h urinary erythritol and IHL content (quantified with 3T Dixon MRI), with adjustment for age, sex, type 2 diabetes, proxies of socioeconomic status/lifestyle, MRI lag time, BMI, and intake of erythritol-containing food items. Second, we performed a genome-wide association study (GWAS) for 24-h urinary erythritol (N = 2,000) and Mendelian randomization (MR) study using common variants in genes associated with erythritol in relation to IHL content. In the fully adjusted model, 24-h urinary erythritol levels were associated with higher (10log) IHL content. The GWAS identified one genome-wide significant locus (rs72686491; TESK2) and replicated two previously reported genetic variants at nominal significance. MR analysis, using these three genetic instruments, did not reveal a statistically significant association between genetically predicted erythritol levels and IHL content. The findings of this population-based study show that higher PPP flux is associated with higher IHL content, a well-established risk factor for type 2 diabetes. An active role for erythritol per se was excluded by MR analysis. ARTICLE HIGHLIGHTS:The role of the pentose phosphate pathway (PPP) in the pathogenesis of intrahepatic lipid (IHL) accumulation has not been studied before in humans. Is the PPP, proxied by 24-h urinary erythritol, associated with IHL content at the population level? In the fully adjusted model, 24-h urinary erythritol levels were associated with higher IHL content. Furthermore, genetically predicted erythritol levels were not associated with liver fat, suggesting that erythritol per se is not responsible for the former observation. Findings of the current study suggest that 1) the PPP is associated with IHL content at the population level, and 2) erythritol could potentially serve as a urinary biomarker for IHL content.

Lifestyle-related factors may affect cognition and increase the risk of developing dementia. However, the neurobiological mechanisms underlying the effect of lifestyle-related factors on the brain are not yet fully understood. Myelin, a vital element of brain health, has previously been shown to be involved in dementia-related cognitive decline. However, whether it plays a role in the relation between lifestyle-related factors and cognitive performance has not been studied. The cerebral myelin content of 4653 participants (51

AIMS:Little is known about how the alignment between daily behaviours, such as physical activity and eating, and chronotype relates to glucose metabolism. We investigated whether alignment of physical activity and meal timing with chronotype was associated with glycaemic parameters and with prediabetes or type 2 diabetes (T2DM). MATERIALS AND METHODS:In a cross-sectional analysis of 1384 participants from The Maastricht Study, we examined associations between behaviour-chronotype alignment and glucose metabolism. Physical activity timing was assessed by accelerometry and defined as the daypart with the highest step count. Meal timing, from a chrono-nutrition questionnaire, was defined as the daypart with the most eating occasions. Chronotype was estimated using the midpoint of sleep on free days corrected for sleep debt. Alignment reflected concordance between behaviour timing and chronotype. Confounder-adjusted logistic and linear regression models estimated associations with (pre)diabetes and with log-transformed fasting plasma glucose (FPG), 2-h post-load glucose (2hPLG), and haemoglobin A1c (HbA1c). RESULTS:Weekday alignment of physical activity timing with chronotype was associated with lower HbA1c (β per 20% more aligned weekdays: -0.48%, 95% CI -0.95, -0.02). Weekday meal timing alignment was associated with lower odds of prediabetes or T2DM (OR aligned vs. misaligned: 0.62, 95% CI 0.43-0.89). No significant associations were observed for weekend alignment, for FPG or 2hPLG, or for interactions between activity and meal timing alignment. CONCLUSIONS:Weekday, but not weekend, alignment of physical activity and meal timing with chronotype was modestly associated with more favourable glucose metabolism. These findings suggest a potential role of behaviour-chronotype alignment in metabolic health, warranting confirmation in prospective and intervention studies.

Abstract Background Little is known about how early food parenting practices (FPPs) influence long-term diet and weight. This study examined associations of FPPs in early childhood— specifically restriction, pressure to eat, monitoring, and stimulation of healthy eating—with dietary intake, BMI, and weight status in young adulthood. Methods We used data from the KOALA Birth Cohort Study in the Netherlands. At age 5 years, mothers reported FPPs using a 5-point Likert scale based on the Child Feeding Questionnaire (scale 1–5). At age 19 years, young adults self-reported their dietary intake and BMI via an online questionnaire, including a food frequency questionnaire. Food intake was categorized as 0–1, 2–3, or 4–7 days per week (scale 1–3). Associations were analyzed using ordinal logistic, binary logistic, linear, and robust regression models. Results Questionnaires at both ages were available for 824 participants. The stimulation FPP “I make sure that my child eats enough healthy food products” was associated with lower unhealthy (B = -0.06), and higher healthy (B = 0.07) food intake. Regarding individual food categories, pressure to eat was associated with consuming more fried snacks (OR = 1.34), and less cooked vegetables (OR = 0.77). Restriction was associated with consuming more fried snacks (OR = 1.41), cookies (OR = 1.34), and pastries (OR = 1.42). Monitoring was associated with higher soda intake (OR = 1.29). The stimulation FPP was associated with consuming less fried snacks (OR = 0.67), cookies (OR = 0.70), and more fruits (OR = 1.65) and cooked vegetables (OR = 1.35). Regarding weight, pressure to eat was associated with lower BMI (B = -0.37) and reduced overweight risk (OR = 0.66), but these associations were nonsignificant after adjusting for BMI z-scores at age 5. Conclusions Early FPPs may shape long-term dietary behaviors. Supportive, structured practices that stimulate healthy eating without coercion appear more beneficial than restrictive or pressuring practices. Longitudinal studies considering mediating and moderating factors are needed to clarify long-term associations between FPPs, dietary, and weight outcomes.

BACKGROUND:n-3 polyunsaturated fatty acids (PUFAs) are linked to lower cardiovascular disease (CVD) risk, potentially via reduced low-grade inflammation (LGI). OBJECTIVE:To examine whether LGI mediates the association of n-3 PUFAs with (pre)clinical CVD, and whether LGI and endothelial dysfunction (ED) mediate the associations with clinical CVD. METHODS:Cross-sectional data from The Maastricht Study were analyzed. Fasted n-3 PUFAs were measured with 1H nuclear magnetic resonance (NMR; n = 3193, 50.8% men, 60 ± 8 years; EDTA plasma) and gas chromatography (GC; n = 1032, 49.6% men, 60 ± 8 years; serum). LGI was defined as the z-score of 6 plasma biomarkers. Preclinical markers of CVD comprised: ED, ankle-brachial index, carotid intima-media thickness, and carotid-femoral pulse wave velocity. CVD was self-reported as coronary heart disease, cerebrovascular disease, or peripheral artery disease. Mediations with n-3 PUFAs (exposures), LGI (mediator), and (pre)clinical CVD (outcomes), and sequential mediations with LGI and ED (mediators) and CVD (outcome) were performed. RESULTS:A total of 522 participants reported CVD. In the NMR subset, n-3 PUFAs were inversely associated with CVD and ED. n-3 PUFAs were inversely associated with LGI (β: -0.09 [-0.12, -0.05]), while LGI was positively associated with ED (β: 0.58 [0.55, 0.61]) and CVD (OR: 1.15 [1.03, 1.28]). LGI mediated the associations with CVD (<5%) and ED (54.4%). LGI and ED mediated small but significant proportions of the associations with CVD (<5%). Similar, although nonsignificant, results were observed in the GC subset. CONCLUSION:Biomarker-based LGI mediated substantial parts of the associations with ED, while together they mediated a small proportion of the associations between n-3 PUFAs and CVD. Further studies are warranted to identify the metabolic pathways via which n-3 PUFAs influence CVD, and the role of LGI in early stages of CVD.

AIMS:Diabetic neuropathy is a frequent complication of type 2 diabetes, yet the relationship between cardiovascular risk factor control and neuropathy remains unclear. We investigated the association between the number of risk factors within guideline-recommended targets and neuropathy in type 2 diabetes. METHODS:In The Maastricht Study, neuropathy was assessed in 3,764 participants (845 with type 2 diabetes; 2,919 without diabetes) using measures of autonomic, sensory, peripheral nerve function, and neuropathic pain. Participants were categorized by the number of eight risk factors within target (HbA1c, blood pressure, BMI, lipids, albuminuria, smoking, physical activity, and diet). Logistic regression estimated odds of neuropathy across strata of risk factors within target. RESULTS:Neuropathy prevalence was 7.8% in individuals with type 2 diabetes and 1.3% in those without diabetes. More risk factors on target were associated with lower odds of neuropathy. Compared with individuals with normal glucose metabolism, adjusted odds ratios were 14.86 (95% CI 7.67-28.82) for 0-2, 3.18 (95% CI 2.01-5.04) for 3-5 and 1.07 (95% CI 0.25-4.56) for 6-8 risk factors on target. CONCLUSIONS:Achieving ≥6 risk factors on target was associated with neuropathy prevalence similar to individuals without diabetes, suggesting comprehensive risk factor control may help prevent neuropathy.

Previous research has revealed associations between diet and health-related quality of life (HRQoL), between diet and kynurenine pathway (KP) metabolites (kynurenines), and between kynurenines and HRQoL in colorectal cancer (CRC) survivors. We examined if kynurenines mediate longitudinal associations between diet and HRQoL in CRC survivors up to 12-months posttreatment. Repeated measurements were performed in 209 stage I-III CRC survivors. Diet was assessed by seven-day dietary records. Plasma kynurenines were analyzed using LC-MS/MS. HRQoL outcomes were assessed through the validated EORTC QLQ-C30. We used confounder-adjusted multilevel parallel-multiple mediator models with all kynurenines simultaneously and single mediator models with established KP ratios to estimate total (c:diet-HRQoL), direct (c':diet-HRQoL), metabolite-specific indirect (ab: diet-metabolite-HRQoL), and total indirect (ab:diet-metabolites-HRQoL) effects. Higher carbohydrate intake was associated with worse role functioning, while higher fiber, alcohol, and zinc intake, and better adherence to the Dutch Healthy Diet (DHD) recommendations were associated with better physical and role functioning (c-path). Associations of fiber and DHD with HRQoL remained statistically significant after controlling for all KP metabolites (c'-path). All kynurenines simultaneously only mediated associations of higher carbohydrate intake with worse role functioning, and of higher alcohol intake with better physical functioning (ab). The kynurenic acid-to-quinolinic acid (KA/QA) ratio and hydroxykynurenine ratio (HKr) significantly mediated associations of carbohydrate, protein, fat, alcohol, magnesium, and zinc intake, and adherence to DHD recommendations, with physical and role functioning (ab). In conclusion, while all kynurenines simultaneously did not mediate diet-HRQoL associations, the KA/QA ratio and HKr did mediate several associations within the first year after CRC treatment.

Dicarbonyls are reactive precursors of advanced glycation end-products. They are formed during food processing, and endogenously in humans during glycolysis and lipid peroxidation. Higher plasma dicarbonyls, particularly methylglyoxal (MGO), promote insulin resistance and type 2 diabetes, but the association between dietary dicarbonyls intake and type 2 diabetes is unknown. This study examined the associations between dietary dicarbonyls and type 2 diabetes incidence. 11,995 incident type 2 diabetes cases and a sub-cohort of 15,797 controls from the prospective multi-center European Prospective Investigation into Cancer and Nutrition (EPIC)-InterAct cohort were included. Intakes of three major dicarbonyls MGO, glyoxal [GO], and 3-deoxyglucosone [3-DG] were estimated at baseline using dietary questionnaires. Type 2 diabetes risk according to dietary dicarbonyl intake was estimated by multivariable-adjusted hazard ratios from Prentice-weighted Cox-regression analyses. Higher intakes of MGO (sample-specific mean intake 3.4 ± 1.3 mg/d) and 3-DG (13.8 ± 10.5) were associated with lower incidence of type 2 diabetes (HR 0.92 [95

Chrono-nutrition, defined as the timing, frequency, and regularity of food intake, may influence glucose metabolism and the risk of type 2 diabetes mellitus (T2DM) through its interaction with circadian rhythms. We examined cross-sectional associations of meal frequency, irregularity, and length of the eating window with glucose metabolism in individuals with normal glucose metabolism, prediabetes, and T2DM. Data were derived from 3,467 participants of a population-based cohort enriched with T2DM. Glucose metabolism status (normal, prediabetes, T2DM) was determined by oral glucose tolerance test and use of glucose-lowering medication. Fasting plasma glucose, 2 h post-load glucose, and HbA1c were measured in venous plasma. Chrono-nutrition was assessed by questionnaire. Multinomial logistic regression estimated odds ratios for prediabetes and T2DM versus normal glucose metabolism. Linear regression models assessed associations with glycemic markers, stratified by glucose metabolism status. Analyses were adjusted for sociodemographic, lifestyle, and clinical factors, including diet quality and caloric intake. A longer eating window was associated with higher odds of prediabetes (OR per hour: 1.07, 95

Background Previous studies on mental disorders have yielded inconsistent findings regarding the relationship between dietary patterns (DPs) and psychological disorders, with most evidence derived from cross-sectional studies. Objectives This study aims to examine the association between major DPs and risk of depression symptoms (DepS) and anxiety disorder (AnxD). Methods This study used data from The Maastricht Study [participants included in analyses for DepS: n = 6967, mean age (SD): 59.94 (8.66) y, female: 49.40%; for AnxD: n = 6634, mean age (SD): 59.94 (8.60) y, female: 49.40%]. A validated food frequency questionnaire was used to assess dietary intakes. DPs were derived using principal component analysis. The Patient Health Questionnaire-9 and the Generalized Anxiety Disorder 7-item questionnaire were used annually to assess DepS and AnxD, respectively. A Cox proportional hazards regression analysis was incorporated to examine the associations. Stratified analyses were performed based on the status of diabetes, sex, body mass index, and smoking. Results Three DPs were identified: “high vegetables and legumes,” “high fast food and sugar,” and “vegetarian-like”. After adjustment for all possible confounders, adherence to the “high fast food and sugar” DP was associated with a higher risk of DepS {hazard ratio [HR] [95% confidence intervals (CIs)] Q5 compared with Q1: 2.13 [1.55, 2.92], P < 0.001, Ptrend < 0.001} and AnxD [HR (95% CIs) Q5 compared with Q1: 2.03 (1.33, 3.10), P = 0.001, Ptrend < 0.001] in the total population. No association was found for other DPs in the total population. A significant interaction was observed between the “vegetarian-like” DP and smoking status on risk of AnxD [interaction term, HR (95% CIs): 0.76 (0.64, 0.90), P = 0.001]. Opposite but nonsignificant associations were noted in never-smokers [HR (95% CIs) Q5 compared with Q1: 1.47 (0.83, 2.60), P = 0.19, Ptrend = 0.06] and former/current smokers [HR (95% CIs) Q5 compared with Q1: 0.62 (0.37, 1.04), P = 0.07, Ptrend = 0.046]. Conclusions A diet high in fast foods, confectionery, fried potatoes, sauces, sugar-sweetened beverages, and refined grains might be associated with an increased risk of depression and anxiety in adults.

STUDY OBJECTIVES:To examine the association of objective and subjective sleep parameters with cognitive functioning and markers of brain morphology. METHODS:This cross-sectional study included 3360 participants (mean age: 59.5 ± 8.5 years; 51.1% female) from The Maastricht Study. Time in bed (TIB) and sleep breaks were objectively estimated using a thigh-worn accelerometer. Subjective sleep continuity was assessed via a single-item, and excessive daytime sleepiness was evaluated using the Epworth Sleepiness Scale. Cognitive testing was administered across three domains-memory, information processing speed, and executive functioning and attention. Markers of brain morphology (e.g. hippocampal volume, gray matter (GM) volume, and white matter volume) were assessed with 3 Tesla magnetic resonance imaging. Linear and logistic regression analyses modeled the associations. RESULTS:A longer TIB (h/night) was associated with worse executive functioning and attention (Blinear = -0.052, 95% CI = -0.084 to -0.019). Categorical analyses showed that a longer TIB (≥9 h/night) was associated with worse executive functioning and attention (B = -0.088, 95% CI = -0.166 to -0.010) compared to a mid-range TIB (≥7 to <9 h/night). A curvilinear association was found between TIB and lower GM volume (Bquadratic = -0.013, 95% CI = -0.025 to -0.001). Sleep breaks (≥2/night) were associated with worse overall cognition (B = -0.069, 95% CI = -0.124 to -0.013), information processing speed (B = -0.125, 95% CI = -0.212 to -0.039), and reduced GM volume (B = -0.068, 95% CI = -0.118 to -0.018). No significant associations were found for memory or other markers of brain morphology. Subjective sleep parameters showed no associations with cognitive functioning or markers of brain morphology. CONCLUSIONS:Adequate and uninterrupted TIB was associated with better cognitive functioning and brain morphology. Statement of Significance With the growing prevalence of dementia, the importance of reducing dementia risk is increasingly acknowledged. Sleep has been identified as a potential risk factor for cognitive decline and dementia, yet studies using objective measures of sleep remain limited. This study explores subjective and objective sleep parameters in relation to cognitive function and brain morphology in a large cohort. The findings show that a long time in bed (TIB) is negatively associated with executive functioning and attention, alongside a curvilinear relationship between TIB and gray matter volume. Additionally, sleep breaks (≥2/night) are linked to lower brain volume and poorer cognition. Future research should investigate longitudinal associations and other sleep aspects, including regularity and sleep disorders, to further our understanding and guide dementia prevention strategies.

BACKGROUND:Cerebral small vessel disease (CSVD) and dementia are increasingly prevalent in the aging population. Diabetes is a major risk factor for CSVD and cognitive impairment, but the underlying mechanisms remain unclear. Methylglyoxal (MGO), a glycolysis by-product and precursor of advanced glycation endproducts (AGEs), is elevated in diabetes and linked to microvascular complications. The aim of this study is to investigate the association between levels of MGO in plasma with MRI-derived markers of CSVD and cognitive impairment in a population-based setting. METHODS:Cross-sectional data from The Maastricht Study, a population-based cohort with an oversampling of type 2 diabetes (age 59.7 ± 8.2 years, 49.9 % male, 26 % type 2 diabetes), were analysed. Plasma MGO concentrations were measured using UPLC-MS/MS after overnight fasting. Participants underwent MRI (n = 1789) and cognitive testing (n = 2515). CSVD markers included white matter hyperintensity volume, lacunar infarcts, and cerebral microbleeds. Logistic and linear regressions, adjusted for confounders, assessed the associations between plasma MGO levels, CSVD markers, and cognitive function. RESULTS:No linear associations were found in any of the outcomes. When dividing the population into tertiles (lowest, intermediate, highest) based on plasma MGO concentration, intermediate compared to lowest levels of plasma MGO were associated with increased odds of having a lacunar infarct (OR: 2.26, 95 %CI: 1.27; 4.01). This was not observed for highest plasma MGO (OR: 1.56, 95 %CI: 0.85; 2.86). CONCLUSION:Intermediate plasma MGO levels were associated with the presence of more lacunar infarcts. Further research is needed to determine whether MGO contributes to lacunar infarct formation in a causal manner.

AIMS:The timing of physical activity may influence metabolic health through interactions with circadian rhythms, yet its role in type 2 diabetes mellitus (T2DM) development is unclear. We investigated associations between time-of-day-specific physical activity and incident T2DM, and whether theoretically reallocating activity from morning to later in the day was associated with changes in T2DM risk. MATERIALS AND METHODS:We included 4615 participants from The Maastricht Study cohort without diabetes (age 59.2 ± 8.6 years; 56.3% women). Device-based physical activity was measured over 7 days using activPAL monitors and classified into light-intensity physical activity (LPA) and moderate-to-vigorous intensity physical activity (MVPA), for morning (06:00-11:59 AM), afternoon (12:00-17:59 PM), evening (18:00-23:59 PM) and night (00:00-05:59 AM). Incident T2DM was assessed during a median 8.2-year follow-up. Cox proportional hazard and isotemporal substitution models were used, adjusted for sociodemographic and lifestyle factors, including diet, employment and sleep duration. RESULTS:During follow-up, 168 participants (3.6%) developed T2DM. Each additional 10 min/day of afternoon LPA or MVPA was associated with lower T2DM risk (LPA: hazard ratio [HR] 0.82, 95% confidence interval [CI] 0.70-0.97; MVPA: HR 0.85, 95% CI 0.72-1.00). Evening MVPA was also inversely associated with T2DM risk (0.65; 0.45-0.93), whereas night-time MVPA was associated with an increased risk (3.64; 1.30-10.17). No significant associations were found of morning LPA and MVPA or evening and night LPA with T2DM incidence. Substitution analyses indicated that reallocating 10 min of morning LPA to afternoon LPA (HR 0.71; 0.54-0.95) or morning MVPA to evening MVPA (HR: 0.64; 0.43-0.96) was associated with a lower T2DM risk, while no other significant associations were observed. CONCLUSIONS:Later-day physical activity, particularly in the afternoon, was associated with a lower incidence of T2DM, independent of intensity. This highlights the potential relevance of activity timing in relation to T2DM incidence.

Objectives: This study investigates the association between dietary intake and ADHD diagnosis and its dimensions in adolescents.Methods: In the KOALA Birth Cohort Study, 810 adolescents aged 16 to 20 years provided information on ADHD diagnosis and completed a food frequency questionnaire. Dietary patterns were extracted using Principal Component Analysis. Parents reported on ADHD symptoms using the Conners' Parent Rating Scale-Revised Short form, and the Impulsivity subscale from the Temperament in Middle Childhood Questionnaire.Results: The 80 adolescents with ADHD scored higher on the Snacking dietary pattern compared to those without ADHD, while they did not differ on Healthy, Animal-based, Sweet, or Beverage dietary patterns. All ADHD symptom scores (Hyperactivity, Inattention and Impulsivity, and ADHD-index) correlated with increased Snacking. Impulsivity was inversely related to Sweet dietary patterns and positively to Beverage dietary patterns.Conclusion: The results highlight the importance of considering ADHD dimensions beyond diagnosis in understanding adolescents' dietary intake.

Elevated methylglyoxal (MGO) levels and altered immune cell responses are observed in diabetes. MGO is thought to modulate immune cell activation. The current study investigated whether fasting or post-glucose-load plasma MGO concentrations are associated with circulating immune cell counts and activation in a large cohort study. 696 participants of The Maastricht Study (age 60.3 ± 8.4 years, 51.9

AIMS:Type 2 diabetes increases the risk of depression, but the mechanisms underlying this association are incompletely understood. We investigated whether microvascular dysfunction, neurodegeneration, low-grade inflammation, advanced glycation end products (AGEs) and arterial stiffness, pathologies that are more common in diabetes, explain, or mediate the association between type 2 diabetes and incident clinically relevant depressive symptoms. MATERIALS AND METHODS:We used prospective data from The Maastricht Study, a population-based cohort study. Diabetes status and potential mediators were assessed at baseline. Clinically relevant depressive symptoms (PHQ-9 score ≥10) were assessed at baseline and each year during a median of 8.1 (IQR 4.2, 10.1) years of follow-up. Mediation analysis was employed to investigate the mediating effect of microvascular dysfunction (retinal, blood and MRI biomarkers), neurodegeneration (retina and MRI biomarkers), low-grade inflammation (blood biomarkers), AGEs (skin and blood biomarkers) and arterial stiffness (tonometry and ultrasound biomarkers). RESULTS:Data of 6091 participants (age, 59.4 years [SD 8.6]; 51.3% women; 23.6% type 2 diabetes) were available. Type 2 diabetes was associated with a higher incidence of clinically relevant depressive symptoms (HR:1.37; 95% CI 1.13, 1.65). This association was partly mediated by microvascular dysfunction (proportion mediated:10.4% [95% CI:3.6%, 17.2%]); neurodegeneration (proportion mediated:12.1% [95% CI: 3.9%, 20.3%]); AGEs (proportion mediated:5.4% [95% CI: 3.0%, 8.8%]); and arterial stiffness (proportion mediated:8.4% [95% CI: 3.3%, 13.5%]); but not by low-grade inflammation. CONCLUSIONS:The association between type 2 diabetes and a higher risk of clinically relevant depressive symptoms is partly mediated by microvascular dysfunction, neurodegeneration, AGEs and arterial stiffness.

OBJECTIVES:Ratio variables (eg, body mass index (BMI), cholesterol ratios, and metabolite ratios) are widely used as exposure variables in epidemiologic studies on cause-and-effect. While statisticians have emphasized the importance of including main effects of the variables that make up a ratio variable in regression models, main effects are still often omitted in practice. The objective of this study is to demonstrate the impact of omitting main effects from regression models with a ratio variable as the focal exposure on bias in the effect estimates and type I error rates. STUDY DESIGN AND SETTING:We demonstrated the impact of omitting main effects in three steps. First, we showed the connection between regression models with ratio variables and regression models with product terms, which are well-understood by epidemiologists. Second, we estimated models with and without main effects of a ratio variable using a real-life data example. Third, we performed a simulation study to demonstrate the impact of omitting main effects on bias and type I error rates. RESULTS:We showed the impact of omitting main effects in regression models with ratio terms. In the real-life example, the ratio term was only statistically significantly associated with the outcome when omitting main effects. The simulation study results indicated that the omission of main effects often leads to biased effect estimates and inflated type I error rates. CONCLUSION:Regression models with a ratio term as an exposure variable need to include main effects to avoid bias in the effect estimates and inflated type I error rates.

The kynurenine pathway (KP) might be involved in pathophysiological processes associated with dementia, but clinical studies reported contradictory results. This systematic review and meta-analysis summarized the available evidence for (i) differences in KP metabolites in patients with cognitive impairment compared to cognitively healthy individuals and (ii) associations between KP metabolites and cognitive functioning. English, full-length articles with prospective, cross-sectional, or case–control study designs, published in Pubmed, Embase, PsychINFO, or the Cochrane Database of Systematic Reviews up to October 2023, were included. Random-effects meta-analyses of standardized mean differences (SMD) were performed. Heterogeneity, meta-regression, small study bias, and study quality assessments were carried out. Of 8797 retrieved studies, 98 were eligible for the systematic review. Meta-analyses comparing Alzheimer’s disease (AD) dementia patients to controls (n = 27 studies) indicated lower CSF levels of tryptophan (SMD = − 0.26 [95

Dietary intake of several macronutrients is associated with plasma kynurenines after colorectal cancer (CRC), and kynurenines have been linked to health-related outcomes. It is unknown how macronutrient substitution affects plasma kynurenines, which may be relevant for developing guidelines to improve post-CRC quality of life through dietary changes. Using iso-caloric substitution models, we investigated how substituting one macronutrient with another is longitudinally associated with plasma tryptophan, kynurenines, and kynurenine ratios in CRC survivors. Measurements were performed at 6-weeks, 6-months, and 12-months post-treatment in 247 stage I-III CRC survivors. Macronutrient intake was measured by 7-d dietary records and plasma kynurenines by LC/MS-MS. For analysis, we applied linear mixed models with false discovery rate (FDR) to adjust for multiple testing. After FDR adjustment, substituting 100 kcal/d of total carbohydrates with 100 kcal/d of total protein was associated with higher plasma concentrations of kynurenic acid (KA), xanthurenic acid (XA), and a higher kynurenic acid-to-quinolinic acid (KA/QA) ratio. Substituting 100 kcal/d of total carbohydrates with 100 kcal/d of total fat was associated with higher tryptophan concentrations, higher KA/QA ratio, and a lower kynurenine-to-tryptophan ratio (KTR) and hydroxykynurenine ratio (HKr). Substituting 100 kcal/d of total fat with 100 kcal/d of total protein was associated with higher XA concentrations. Altogether, iso-caloric macronutrient substitutions, particularly substituting carbohydrates with protein or fat, were longitudinally associated with higher concentrations of potentially favourable kynurenines and ratios (i.e., KA, XA, and KA/QA ratio) and lower ratios with pro-inflammatory or neurotoxic properties (i.e., KTR and HKr) in CRC survivors up to 12-months post-treatment.