Graduation Year

2023

Document Type

Dissertation

Degree

Ph.D.

Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department

Public Health

Major Professor

Amy C. Alman, Ph.D.

Co-Major Professor

Heewon L. Gray, Ph.D., RDN

Committee Member

Getachew A. Dagne, Ph.D.

Committee Member

Arpita Basu, Ph.D., MPH, RD/LD

Keywords

mixed-effects model, healthy eating, Mediterranean diet, microbiome, ultra-processed food

Abstract

Introduction: The prevalence of type 1 diabetes mellitus (T1DM) has been increasing drastically in the United States. People with T1DM are at a higher risk of developing diabetes-related complications, including cardiovascular disease (CVD). Diet is one of the most modifiable factors and healthy eating could reduce CVD risk among individuals with T1DM. The current 2015-2020 Dietary Guidelines for Americans focus on overall healthy eating patterns rather than individual foods and nutrients. A healthy dietary pattern or high-quality diet can be measured by diet quality indexes, and generally contains minimal ultra-processed foods (UPF). Although substantial evidence has linked poor diet quality and high UPF intake with negative health outcomes, the impact of dietary intake in people with T1DM and their associations with cardiovascular risk factors and glycemic status remains understudied. In addition, oral microbiome plays a crucial role in human health and oral microbiota dysbiosis has been associated with atherosclerosis and CVD. Diet may shape oral microbial profile and affect CVD risk through diet-microbiome-host interaction. No study has investigated the associations among diet, cardiometabolic profile, and salivary microbiome in people with T1DM.

Objective: The objectives of this study were to investigate diet quality and UPF consumption among people with and without T1DM, and to examine their associations with cardiovascular biomarkers and salivary microbiome, using data collected from a longitudinal Coronary Artery Calcification in Type 1 Diabetes (CACTI) cohort study.

Methods: A total of 1,416 adults (N=652 T1DM, N=764 non-diabetic controls) aged 19 to 56 years with no known history of coronary heart disease were enrolled. Dietary intake was assessed using a self-administered Harvard food frequency questionnaire, and cardiovascular biomarkers were measured at baseline, 3-, 6-, and 14-year follow-ups. The salivary microbiome was only collected from a subset of the cohort at the last follow-up visit (N=109 T1DM and 99 non-diabetic controls). Alternative Healthy Eating Index-2010 (AHEI-2010) and Mediterranean-style Dietary Pattern Score (MSDPS) were used to compute the dietary quality scores, and NOVA food classification system was used to compute UPF consumption among participants during all four visits. Linear mixed-effects models were used to estimate the prospective associations of dietary quality and UPFs, with cardiovascular biomarkers, adjusting for potential confounders. Cross-sectional associations of UPF intake with salivary microbiome, and salivary microbiome with cardiovascular biomarkers, were examined using Microbiome Multivariable Association with Linear Models (MaAsLin2). Statistical analyses were performed in SAS version 9.4 and R 3.6.0.

Results: People with T1DM had lower but non-significant diet quality scores (AHEI-2010: 40.8 ± 10.4 in T1DM, 41.6 ± 10.3 in non-diabetic controls; MSDPS: 24.9 ± 7.8 in T1DM, 25.3 ± 7.7 in non-diabetic controls) and higher UPF consumption (T1DM: 8.7±7.8 servings/d, controls: 7.3±6.7 servings/d, all P<0.05) compared to non-diabetic controls at baseline. Participants with T1DM had significantly higher consumption of soft drinks, red or processed meat, sweets, savory snacks, sodium, and lower fish and other seafood than the controls at one or multiple time points (all P<0.05). Diet quality was significantly associated with better cardiometabolic profiles, including lower BMI (AHEI-2010: β=-0.19±0.05, MSDPS: β=-0.14±0.06), waist circumference (AHEI-2010: β=-0.49±0.14, MSDPS: β=-0.44±0.16), systolic blood pressure (AHEI-2010: β=-0.60±0.22, MSDPS: β=-1.31±0.26), and higher HDL-C (AHEI-2010: β=1.25±0.24, MSDPS: β=0.91±0.28) (all P<0.05), after adjusting for a wide range of potential confounders including diabetes stauts. Higher UPF consumption was prospectively associated with worse cardiometabolic profiles, including higher BMI (β=0.10±0.04), waist circumference (β=0.30±0.11), systolic blood pressure (β=0.55±0.19), triglycerides (β=1.79±0.84), and lower HDL-C (β=-0.43±0.20) (all P<0.05), after adjusting diabetes status. There was no statistically significant association of salivary microbial alpha diversity with UPFs and cardiovascular risk factors. However, consumption of UPFs was positively associated with the genera Aggregatibacter (β= 0.61±0.14), Leptotrichia (β=0.45±0.14), and Proteobacteria members (β=0.30±0.07) (all FDR-adjusted p-value: q<0.05).

Conclusions: An overall healthy dietary pattern is beneficial for reducing CVD risk, especially for T1DM who are at a greater risk of developing diabetes-related complications. People with and without T1DM are encouraged to replace UPFs with whole foods, lower the consumption of soft drinks, red and processed meat, sweets, and sodium, and increase the consumption of fruits, vegetables, whole grains, and fish and other seafood, to improve overall diet quality and better protect against CVD risk. The findings provided new insights into the independent associations between healthy eating and cardiovascular disease risk among people with and without T1DM. The study calls for the need to develop targeted nutrition interventions promoting healthful eating patterns to reduce the risk of cardiovascular disease among people with T1DM. Future randomized controlled trials or longitudinal studies with large sample sizes and diverse racial/ethnic groups are needed to assess trends in dietary quality and study the impact of cultural and food preference attributes on dietary pattern adherence among people with T1DM.

Download

Share

COinS