The mean intake for total dietary fiber was 13.96 g/d, and mean intakes for soluble and insoluble fiber were 3.71 g/d and 10.16 g/d, respectively. gives means, standard deviations (SD) and proportions of participant characteristics across total dietary fiber. Based on the descriptive statistics, there were no significant differences in participant characteristics across tertiles of total dietary fiber intake, except for race/ethnicity, education, vitamin E supplementation, arthritis and hypertension. The majority of participants identified themselves as non-Hispanic whites, ranging from 49% to 70% across tertile of total dietary fiber (p <0.001). HEAL participants were well educated as 74.3% reported having attended college or graduate school. Vitamin E supplementation varied across the tertile of total dietary fiber intake, with 50.0% taking supplements in tertile one, 64.8% in tertile two and 67.0% within tertile three (p = 0.001). Rates of arthritis were found to differ across the tertile of dietary fiber (p = 0.04), as well as hypertension (p = 0.02)
Anthropometric and lifestyle characteristics by tertiles of total dietary fiber levels among 698 female breast cancer survivors
The overall mean serum CRP was 3.32 mg/L with significant differences across the fiber intake distribution (). Mean serum CRP concentrations were lowest among participants with the highest grouping (third tertile) of total dietary fiber intake, 2.93 mg/L; whereas those in the second and first fiber intake tertiles had a mean CRP concentration of 3.24 mg/L and 3.79 mg/L, respectively (p <0.036). SAA concentrations did not differ across fiber distribution. Among those with the lowest intake of total dietary fiber, the mean total dietary fiber intake was 7.56 g/day, and correspondingly the mean intakes for soluble and insoluble fiber were 2.05 g/day and 5.45 g/day. For those classified into tertile two, the mean total dietary fiber intake was 12.95 g/day, with mean soluble fiber intake of 3.43 g/day and mean insoluble fiber intake of 9.43 g/day. Women classified into the highest intake of dietary fiber (tertile three), the mean total dietary fiber intake was 21.35 g/day, soluble fiber intake was 5.64 g/day, and insoluble fiber intake was 15.48 g/day.
Inflammatory markers and dietary intake by tertiles of dietary fiber intake among 689 female breast cancer survivors
Linear regression models tested associations of total, soluble, and insoluble dietary fiber intake with both CRP and SAA (). We constructed a multivariate-adjusted model (adjusting for age, BMI, race/ethnicity/site, and other known moderators of CRP, including current smoking status, postmenopausal status, physical activity, alcohol consumption, dietary fat intake, total energy, number of cardiovascular and inflammatory related conditions, medication use, disease stage and adjuvant treatment used), in which a strong inverse association of total dietary fiber with serum CRP was observed (β, −0.029; 95% CI, −0.049, −0.008; p = 0.006). We observed no associations between total dietary fiber intake and SAA.
Regression coefficients from analysis of covariance using total dietary fiber intake, soluble dietary fiber intake, and insoluble dietary fiber intake level as the primary independent variable and CRP and SAA as the dependent variable.
We next examined the relationship between fiber types, soluble and insoluble, and measures of inflammation (). Insoluble fiber was strongly, inversely associated with CRP concentrations (multivariate-adjusted model: β, −0.039; 95% CI, −0.064, −0.013; p = 0.003). However, no relationship was observed between soluble fiber and CRP concentrations. No associations were found for either total fiber or any fraction with serum SAA concentrations.
The odds of elevated CRP, defined as having a serum CRP concentration ≥3 mg/L, was examined. For these analyses, participants in the highest two tertiles of total, soluble and insoluble dietary fiber intake were compared to participants in the lowest tertile of corresponding fraction of dietary fiber intake (). While there was a suggestion that total dietary fiber intake may protect against elevated CRP, the association did not reach statistically significance in the multivariate-adjusted model (OR, 0.57; 95% CI, 0.30, 1.07). Soluble dietary fiber was also inversely related to elevated CRP, but the association was not statistically significant (OR, 0.86; 95% CI, 0.45, 1.63). However, participants in the highest tertile of insoluble dietary fiber intake had a 49% reduced likelihood of having elevated CRP when compared to the lowest tertile (OR, 0.51; 95% CI, 0.27, 0.95, p for trend = 0.044).
Odds ratios of elevated CRP (>3.00 mg/l) risk comparing across tertile of total, soluble and insoluble dietary fiber intake, HEAL study.
presents mean, SD, median, and interquartile range (IQR) for serum concentrations of CRP and SAA by racial/ethnic groups. Serum CRP concentrations were highest among non-Hispanic black women, (4.46 mg/l), followed by Hispanics (3.60 mg/l), and non-Hispanic white women (2.83 mg/l, p <0.001). Concentrations of SAA were similar across racial/ethnic groups. In multivariate linear regression analysis, non-Hispanic white women had an inverse association of both total dietary fiber intake (β, −0.031; 95% CI, −0.057, −0.006) and insoluble fiber intake (β, −0.042; 95% CI, −0.074, −0.010) with circulating concentrations of CRP. For every one gram increase in total and insoluble fiber intake, one may expect to see an approximate 3% and 4% decrease in serum CRP concentration (mg/L), respectively. A similar relationship was seen among non-Hispanic black women, but the associations were not statistically significant. No associations were observed between total, soluble, and insoluble dietary fiber intake and CRP among Hispanic women.
TABLE 5 Distribution of Inflammatory Markers and Age-adjusted regression coefficients (β) from analysis of covariance between total dietary fiber intake, soluble dietary fiber intake, and insoluble dietary fiber intake level with CRP and SAA by race/ethnicity (more ...)