Association between Carbohydrate Intake and Serum Lipids

Yunsheng Ma, Youfu Li, David E. Chiriboga, Barbara C. Olendzki, James R. Hébert, Wenjun Li, Katherine Leung, A Hafner, Ira S. OckeneJournal of the American College of Nutrition 2006 · DOI

TL;DR

This study investigated the complex relationship between dietary carbohydrate intake and serum lipid levels in healthy adults. It found that higher intake of processed carbohydrates, particularly those with a high glycemic index (GI) and glycemic load (GL), is associated with less favorable lipid profiles, potentially increasing the risk of metabolic syndrome, diabetes, and heart disease.

What Problem Does This Paper Address?

Cardiovascular disease (CVD) remains a major health concern, and diet is a well-established modifiable risk factor. While the role of saturated fat and cholesterol in CVD has been extensively studied, the impact of dietary carbohydrates (CHO) on blood lipids and CVD risk has received less attention, despite a growing trend towards lower-carbohydrate diets for weight loss. Previous research has suggested that carbohydrate quality, as measured by glycemic index (GI) and glycemic load (GL), might influence lipid profiles, but the long-term effects and the interplay between different carbohydrate metrics remain unclear. This paper aims to clarify these associations by examining both cross-sectional and longitudinal data on carbohydrate intake and serum lipids.

Methods

The researchers utilized data from the Seasonal Variation in Blood Cholesterol Levels (SEASONS) study, a prospective observational study. They collected blood samples and conducted 24-hour dietary recalls quarterly for five consecutive quarters from 574 healthy adults.

Key methodological aspects include:

  • Dietary Assessment: Serial 24-hour dietary recalls were used to capture dietary changes over time. This method is considered more accurate than food frequency questionnaires for assessing nutrient intake.
  • Glycemic Index (GI) and Glycemic Load (GL) Calculation: GI and GL were computed for individual food items and then aggregated to represent daily intake. The GI ranks carbohydrates based on their glycemic effect, while GL considers both the GI and the amount of carbohydrate consumed.
  • Statistical Modeling: Linear mixed models were employed to analyze the relationships between carbohydrate intake variables (total CHO, % calories from CHO, GI, GL) and serum lipids (total cholesterol, LDL-C, HDL-C, triacylglycerols). These models accounted for within-subject correlations and included adjustments for various confounding factors such as age, gender, BMI, physical activity, and other dietary components. Both cross-sectional (between-subject) and longitudinal (within-subject) effects were analyzed.

Datasets

The study used data from the Seasonal Variation in Blood Cholesterol Levels (SEASONS) study. This dataset comprised 574 healthy adults from central Massachusetts, who were followed prospectively for one year. Data collected included:

  • Quarterly blood samples for serum lipid analysis.
  • Quarterly 24-hour dietary recalls.
  • Information on physical activity, demographic variables, and other potential confounders.

No new benchmarks were introduced; the study utilized existing methodologies for GI/GL calculation and statistical analysis.

Key Insights and Ideas

  • Complex and Unfavorable Effects of Processed Carbohydrates: The study highlights that increased intake of highly processed carbohydrates, particularly those with high GI and GL, is associated with a less favorable lipid profile. This suggests that not all carbohydrates are created equal when it comes to cardiovascular health.
  • Inverse Relationship with HDL-C and Triacylglycerols: Higher carbohydrate intake, especially when measured by % of calories from CHO, was linked to lower HDL-C (the "good" cholesterol) and higher triacylglycerols. This combination is a known risk factor for metabolic syndrome and heart disease.
  • Potential Benefit of Lower Total/LDL-C, but with a Caveat: While some measures of carbohydrate intake (total CHO, GL) were associated with lower total and LDL-C (the "bad" cholesterol) cross-sectionally, this effect was offset by a greater reduction in HDL-C, leading to a potentially unfavorable increase in the TC:HDL-C ratio.
  • Longitudinal vs. Cross-Sectional Findings: Longitudinal analyses revealed fewer significant associations compared to cross-sectional analyses. This suggests that the impact of carbohydrate intake on lipids might be more pronounced when comparing different individuals (cross-sectional) rather than tracking changes within individuals over time (longitudinal), possibly due to smaller changes in diet and lipids over the one-year study period.
  • Independent Effects of CHO and GI: Preliminary analysis suggested that carbohydrate quantity (grams or % calories) and quality (GI) might have independent effects on blood lipids.

Key Results

  • Cross-sectional findings:
    • Higher total carbohydrate intake, % calories from carbohydrate, GI, and GL were associated with lower HDL-C and higher triacylglycerols.
  • Higher total carbohydrate intake and GL were associated with lower total and LDL-C.
  • Longitudinal findings (after adjustment for covariates):
    • GL was positively associated with total cholesterol and LDL-C.
  • % of calories from carbohydrate was inversely associated with HDL-C.
  • TC/HDL-C Ratio: % of calories from carbohydrate was associated with a higher TC/HDL-C ratio, indicating a less favorable lipid profile.
  • Gender Differences: Males had higher LDL-C, triacylglycerols, and caloric intake, while females had higher HDL-C and % of calories from carbohydrates.

Limitations and Future Work

The authors acknowledge several limitations:

  • Self-Reported Data: Diet and physical activity data were self-reported, which can be subject to information bias.
  • Generalizability: The study population was predominantly white, well-educated, and employed, which may limit the generalizability of findings to other demographic groups.
  • Causality: The cross-sectional associations, while consistent, do not establish causality.
  • Short Follow-up: The one-year follow-up period might not be long enough to observe significant longitudinal changes in lipids related to dietary shifts.

Future work suggested by the authors includes:

  • Randomized Controlled Trials (RCTs): The authors strongly recommend RCTs to investigate these associations and determine the implications for lipid management. This would allow for stronger causal inferences.
  • Long-term Intervention Studies: Further long-term studies in diverse populations are needed to clarify the relationship between carbohydrate factors, other dietary aspects, and physical activity in relation to lipid profiles and CVD.

Summary updated

Abstract

Results suggest that there is a complex and predominantly unfavorable effect of increased intake of highly processed carbohydrate on lipid profile, which may have implications for metabolic syndrome, diabetes, and coronary heart disease. Further studies in the form of randomized controlled trials are required to investigate these associations and determine the implications for lipid management.
BibTeX
@article{ma2006association,
  author = {Yunsheng Ma and Youfu Li and David E. Chiriboga and Barbara C. Olendzki and James R. Hébert and Wenjun Li and Katherine Leung and A Hafner and Ira S. Ockene},
  title = {{Association between Carbohydrate Intake and Serum Lipids}},
  journal = {Journal of the American College of Nutrition},
  year = {2006},
  doi = {10.1080/07315724.2006.10719527},
}
Figure