Abstract
Purpose:
To prospectively examine the relation between adolescent dietary intake and cardiometabolic risk (CMR) clustering at the end of adolescence.
Methods:
Data from the NHLBI Growth and Health Study on 1369 girls enrolled at ages 9–10 in 1987–1988 and followed for 10 years were used to estimate the relative risk of having multiple (≥ 2 or ≥ 3) risk factors in late adolescence associated with usual food intake patterns from 9–17 years of age. Mean food intakes were derived from multiple 3-day diet records and CMR factors included larger waist circumference, insulin resistance, low high-density lipoprotein-cholesterol, high low-density lipoprotein-cholesterol, high triglycerides, and elevated systolic or diastolic blood pressures.
Results:
Of 1,369 subjects, 18.4% girls had 3–6 prevalent risk factors by the end of adolescence and 35.0% had at least two. Higher intakes of fruit and non-starchy vegetables, dairy, and grains were independently associated with having fewer risk factors as were eating patterns characterized by higher combined intakes of these food groups. After adjusting for age, race, socio-economic status, height, physical activity, and television watching, girls with high intakes of dairy and fruits and non-starchy vegetables (vs. those with lower intakes of both) were nearly 50% less likely to have three or more CMR factors in late adolescence; girls with higher intakes of grains plus fruits and non-starchy vegetables were nearly 60% less likely.
Conclusion:
These results suggest that healthy food consumption patterns during adolescence may prevent accumulation of cardiometabolic risk.
Keywords: Nutrition, central obesity, blood pressure, lipids, insulin resistance, youth
Introduction
While the metabolic syndrome as a diagnostic construct has come under question, clustering of cardiovascular risk factors continues to serve as the basis for adult cardiovascular risk assessment. The roots of cardiovascular disease are established early and cardiovascular risk clustering persists from early childhood through adolescence and adulthood [1;2]. The presence of multiple cardiovascular risk factors in younger populations is associated with concurrent vascular dysfunction and therefore may serve as an important tool for identifying adolescents at higher cardiovascular risk [3].
The role of diet patterns in the development of cardiometabolic risk (CMR) in adults has been examined in a number of studies [4–6]. These studies have explored the impact of specific food groups as well as related patterns of intake of foods, macronutrients and other nutrients [7]. In contrast, the role of food-based consumption in the development of CMR clustering amongst children has been infrequently studied.[8] Previous analyses of data from the National Growth and Health Study (NGHS) have examined the effects of dietary patterns on separate adolescent health outcomes including weight, blood pressure, and lipid levels but the effect of dietary intakes on the clustering of CMR factors is unknown [9–11].
The goal of the current analysis was to examine the effects of usual food group intakes in earlier adolescence on the clustering of CMR factors in later adolescence.
Methods
The analyses were conducted using data from the National Heart, Lung and Blood Institute’s Growth and Health Study (NGHS), a longitudinal study of the development of obesity and other cardiovascular-related outcomes in adolescent girls. Beginning in 1987–1988, the NGHS enrolled 2,379 subjects at 9–10 years of age from three representative urban and suburban clinical sites. Approximately equal numbers of blacks and whites were enrolled; subjects were followed annually until ages 19–20. Details of the study design have been previously published [12]. The current analyses were approved by the Institutional Review Board of Boston University and include 1,369 girls with complete data on diet and six CMR factors described below.
Dietary Intake
Diet was assessed using 3-day diet records during eight of the 10 study years. Data were entered into the Nutrition Data System (NDS) nutrient calculation program (of the Nutrition Coordinating Center of the University of Minnesota) [13] using version 19 food tables (the version most appropriate to the period of data collection). Each subject’s food intake was calculated in Food Pyramid servings by the investigators at Boston University through linking NDS food codes with those from the USDA’s “MyPyramid Equivalents Database, Version 2.0 for USDA Survey Food Codes”[14]. The final Food Pyramid data set contains each subject’s intake in the five major food groups (i.e., fruit, vegetables, dairy, meat/other proteins, grains) and all subgroups.
Measurement of CMR Factors
The following factors related to CMR were considered: waist circumference, blood pressure, low-density lipoprotein (LDL) cholesterol, high-density lipoprotein (HDL) cholesterol, triglycerides (TG), and insulin resistance (IR). Mean values for each risk factor at the final exam served as primary outcomes. Waist circumference was measured in duplicate at the narrowest part of the torso against the skin. Blood pressure was measured by standardized protocol with a mercury sphygmomanometer. Fasting lipids and glucose were analyzed at a Johns Hopkins laboratory (a participant in CDC/NHLBI Lipid Standardization Program). LDL cholesterol was calculated using a modified Friedewald equation: [LDL] = [(TC)] – [HDC] – [TG]/6.5[15]. IR was estimated using the homeostasis model assessment (HOMA-IR)[16].
Potential Confounders
Potential confounders included age, race, socio-economic status (SES), height, physical activity, and television/video viewing time. Girls were self-described as “black” or “white”. Annual household income and education were used to classify SES as low, moderate or high. Physical activity was assessed during eight years using a validated Health Activity Questionnaire (HAQ) [17]. Usual ours spent watching television/video was assessed annually by questionnaire. There was no confounding by energy, macronutrient intakes, or family medical history variables so these variables were dropped from the final models.
Statistical Analysis
Usual dietary intake was estimated as a mean of all-available diet records from ages 9 to 17 years. Up to eight sets of 3-day diet records were collected between 9 and 17 years of age. On average, the girls completed 18 days of diet records. The mean intake was used to derive a more stable estimate of usual intake patterns and to minimize the effects of measurement error, seasonal variability in intake, and changes in consumption linked with periods of growth. The clustering of six CMR factors in late adolescence (ages18–20) was examined in relation to selected diet patterns based on usual intake over the course of the study. Each girl was classified as being at elevated risk (or not) for each of six factors as follows: waist circumference (≥88 cm), systolic and/or diastolic blood pressure (≥90th percentile for age, sex and height), LDL (≥110 mg/dL), HDL (<50 mg/dL), serum TG (≥110mg/dl), and HOMA-IR (≥4). Cutoff values were selected from several sources. “At risk” waist circumference and TG were derived from the National Cholesterol Education Program (NCEP) guidelines [18]. Low HDL followed the International Diabetes Federation guidelines for late adolescent girls [19]. “At risk” blood pressure was defined using the National High Blood Pressure Education Program Working Group on High Blood Pressure in Children and Adolescence [20]. The elevated LDL criterion was based on the NCEP classification of “borderline high” levels for children and adolescents [21]. While there is no single accepted value for defining insulin resistance using HOMA-IR among adolescents, we chose a cutoff value of 4.0 based on previously-published studies as well as the data distribution in the current study [22]. Subjects received a score of one for each risk factor in which their value fell in the at-risk range or a zero otherwise. The sum of the six risk factors served as an overall risk clustering score.
Multiple logistic regression analysis was used to estimate the relative risk (RR) of having multiple (≥ 2 or ≥ 3) risk factors in later adolescence (18–20 years of age) associated with selected diet patterns including dairy, fruits and non-starchy vegetables (FnsV), and grains. Total dairy is presented in the results since no differences were observed in effects of low-fat vs. fullfat dairy. Patterns of intake were explored for two-way combinations of the above food groups, with intake of each food dichotomized (e.g., <2 vs. ≥ 2 cup-equivalents for dairy and <1.5 vs. ≥ 1.5 cup-equivalents per day for FnsV). In this way, four mutually-exclusive categories were derived (e.g., (a) low intakes of both dairy and FnsV, (b) low intake of dairy and high intake of FnsV, (c) high intake of dairy but low intake of FnsV, and (d) high intakes of both dairy and FnsV). Other diet patterns combined dairy and grain consumption as well as FnsV and grains. Since whole grain consumption was very low in this cohort, total grains were analyzed instead. Cutpoints were based on distribution of levels of intake to optimize analytic power.
Results
Table 1 provides descriptive data on subjects according to three food groups—dairy, FnsV, and grains. Girls reporting the highest intakes of dairy or FnsV were more likely to be white. They also had a lower mean BMI, were more active and watched less television than girls with lower intakes in either of these food groups. Somewhat similar, but weaker, patterns were found for higher intakes of grains. Dairy intakes were directly associated with energy-adjusted protein intakes and inversely associated with the percent of calories from fat.
Table 1.
Subject characteristics according to usual mean daily dietary intakes at ages 9–17 years
| Dairy Servingsa | Fruit & Non-Starchy Vegetable Servingsa | Grains Servingsa | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Subject | <1.0 | 1.0–<2.5 | >2.5 | <1.0 | 1.0–<1.5 | >1.5 | <5.0 | 5.0–<7.0 | >7.0 |
| Characteristics | (n=223) | (n=959) | (n=187) | (n=392) | (n=449) | (n=528) | (n=299) | (n=721) | (n=349) |
| N (column percent) | |||||||||
| Race | |||||||||
| White | 45 (20.2%) | 440 (45.9%) | 154 (82.4%) | 166 (42.3%) | 186 (41.4%) | 287 (54.4%) | 115 (38.5%) | 357 (49.5%) | 167 (47.9%) |
| Black | 178 (79.8%) | 519 (54.1%) | 33 (17.6%) | 226 (57.7%) | 263 (58.6%) | 241 (45.6%) | 184 (61.5%) | 364 (50.5%) | 182 (52.1%) |
| SES | |||||||||
| Low | 44 (19.7%) | 221 (23.1%) | 28 (15.0%) | 113 (28.8%) | 115 (25.6%) | 65 (12.3%) | 68 (22.7%) | 141 (19.6%) | 84 (24.1%) |
| Moderate | 109 (48.9%) | 403 (42.0%) | 63 (33.7%) | 182 (46.4%) | 197 (43.9%) | 196 (37.1%) | 137 (45.8%) | 306 (42.4%) | 132 (37.8%) |
| High | 70 (31.4%) | 335 (34.9%) | 96 (51.3%) | 97 (24.8%) | 137 (30.5%) | 267 (50.6%) | 94 (31.5%) | 274 (38.0%) | 133 (38.1%) |
| Mean (s.d.)b | |||||||||
| BMI (kg/m2) | 23.1 (5.4) | 22.0 (4.8) | 20.8 (4.2) | 22.5 (5.1) | 22.0 (4.6) | 21.7 (4.8) | 23.0 (5.1) | 21.9 (4.9) | 21.3 (4.3) |
| Activity (METS) | 17.4 (9.2) | 19.6 (9.8) | 22.6 (11.0) | 17.4 (9.1) | 18.9 (9.5) | 22.0 (10.5) | 18.7 (8.9) | 19.9 (10.0) | 20.1 (10.8) |
| Television (hrs/day) | 5.3 (1.9) | 4.7 (2.1) | 3.4 (2.1) | 5.0 (2.0) | 4.9 (2.0) | 4.1 (2.2) | 4.9 (2.0) | 4.5 (2.0) | 4.6 (2.3) |
| Total energy (kcals/day) | 1745 (331) | 1889 (361) | 2099 (374) | 1751 (320) | 1897 (364) | 1998 (376) | 1565 (255) | 1876 (289) | 2214 (338) |
| % kcals-protein | 13.7 (2.0) | 14.0 (1.8) | 14.9 (1.9) | 14.1 (2.05) | 14.0 (1.9) | 14.2 (1.1) | 14.6 (2.2) | 14.0 (1.8) | 13.8 (1.7) |
| % kcals-carbohydrates | 51.2 (5.1) | 51.8 (4.9) | 52.1 (4.9) | 50.0 (4.9) | 51.2 (4.3) | 53.4 (5.0) | 50.8 (5.2) | 51.8 (4.7) | 52.3 (5.1) |
| % kcals-fat | 36.0 (4.0) | 35.2 (3.9) | 34.2 (4.2) | 36.7 (3.7) | 35.7 (3.5) | 33.7 (4.1) | 35.6 (4.1) | 35.1 (3.8) | 35.0 (4.4) |
| Fruit | 0.83 (0.6) | 0.89 (0.6) | 1.1 (0.7) | 0.37 (0.2) | 0.72 (0.2) | 1.5 (0.6) | 0.80 (0.6) | 0.89 (0.6) | 1.0 (0.7) |
| Vegetables | 1.1 (0.5) | 1.0 (0.4) | 1.0 (0.4) | 0.84 (0.3) | 1.1 (0.4) | 1.2 (0.5) | 0.94 (0.4) | 1.0 (0.4) | 1.1 (0.4) |
| Grains | 5.4 (1.2) | 6.2 (1.4) | 6.8 (1.6) | 5.7 (1.2) | 6.1 (1.4) | 6.5 (1.5) | 4.3 (0.5) | 6.0 (0.6) | 8.0 (1.0) |
| Dairy products | 0.78 (0.2) | 1.6 (0.4) | 3.0 (0.5) | 1.6 (0.7) | 1.6 (0.7) | 1.8 (0.8) | 1.4 (0.6) | 1.7 (0.7) | 1.9 (0.7) |
| Meat products | 5.0 (1.7) | 4.6 (1.6) | 4.3 (1.6) | 4.4 (1.4) | 4.7 (1.6) | 4.8 (1.7) | 4.3 (1.4) | 4.5 (1.5) | 5.1 (1.8) |
SES = socioeconomic status; BMI = body mass index; METS = metabolic equivalents derived from physical activity scores
Dairy and fruit & non-starchy vegetables expressed as cup equivalents (i.e., for fruit & non-starchy vegetables, 1 cup raw or cooked non-starchy vegetable or fruit, ½ cup dried vegetable or fruit, or 1 cup (8 fl. oz.) 100% vegetable or fruit juice, or 2 cups leafy salad greens); grain intakes expressed as ounce equivalents (i.e. amount equivalent to 1 bread slice,1 cup cereal, or 1/2 cup cooked pasta/rice)
Mean values from ages 9–17 years
The first row in Table 2 shows that of the 1,369 girls at the end of follow up, 30.1% had no CMR factors while 34.9% had only one, 16.6% had two and 18.4% had three or more. While the proportion of girls with clustering of risk factors was very similar in whites and blacks, the types of risk factors observed were different (data not shown). Increased waist size, elevated BP and insulin resistance were twice as common in blacks as in whites. In contrast, abnormal lipid levels were more common in whites than blacks; in fact, 27.5% of white girls had at least two lipid related abnormalities compared with 18.2% of black girls. In particular, white girls were more than twice as likely to have elevated triglyceride levels. Table 2 also shows the degree of CMR clustering according to dietary intakes in three food groups. Unfortunately, the distribution of FnsV intake among these girls did not allow us to evaluate particularly high intakes of these foods. Girls consuming 1.5 cups of FnsV per day had fewer risk factors than girls consuming less. In contrast, 23.3% of girls consuming less than one cup per day had 3 or more prevalent risk factors by late adolescence. There were similar but weaker trends for intakes of dairy and grains.
Table 2.
Clustering of cardiometabolic risk factors at 18–20 years of age overall and by categories of dietary intakes
| Number of Risk Factorsa | ||||||
|---|---|---|---|---|---|---|
| N | 0 | 1 | 2 | 3 | 4–6 | |
| N (row percent) | ||||||
| Overall | 1369 | 412 (30.1%) | 478 (34.9%) | 227 (16.6%) | 152 (11.1%) | 100 (7.3%) |
| Whites | 639 | 187 (29.3%) | 242 (37.9%) | 103 (16.1%) | 76 (11.9%) | 31 (4.9%) |
| Blacks | 730 | 225 (30.8%) | 236 (32.3%) | 124 (17.0%) | 76 (10.4%) | 69 (9.5%) |
| Dairy (cup-eq/day)b | ||||||
| <1.0 | 223 | 67 (30.0%) | 75 (33.6%) | 38 (17.0%) | 28 (12.6%) | 15 (6.7%) |
| 1.0–<2.5 | 959 | 283 (29.5%) | 333 (34.7%) | 165 (17.2%) | 98 (10.2%) | 80 (8.3%) |
| >2.5 | 187 | 62 (33.2%) | 70 (37.4%) | 24 (12.8%) | 26 (13.9%) | 5 (2.7%) |
| p-value for χ2 | 0.1200 | |||||
| FnsV (cup-eq/dayb | ||||||
| <1 | 392 | 104 (26.5%) | 127 (32.4%) | 70 (17.9%) | 52 (13.3%) | 39 (10.0%) |
| 1–<1.5 | 449 | 133 (29.6%) | 161 (35.9%) | 74 (16.5%) | 51 (11.4%) | 30 (6.7%) |
| >1.5 | 528 | 175 (33.1%) | 190 (36.0%) | 83 (15.7%) | 49 (9.3%) | 31 (5.9%) |
| p-value for χ2 | <0.001 | |||||
| Grains (oz-eq/day)b | ||||||
| <5 | 299 | 90 (30.1%) | 97 (32.4%) | 61 (20.4%) | 27 (9.0%) | 24 (8.0%) |
| 5–<7 | 721 | 211 (29.3%) | 248 (34.4%) | 109 (15.1%) | 90 (12.5%) | 63 (8.7%) |
| >7 | 349 | 111 (31.8%) | 133 (38.1%) | 57 (16.3%) | 35 (10.0%) | 13 (3.7%) |
| p-value for χ2 | <0.0960 | |||||
FnsV = fruit and non-starchy vegetables; BP = blood pressure; HDL = high-density lipoprotein: LDL = low-density lipoprotein; TG = triglycerides; HOMA-IR = homeostasis model assessment-insulin resistance
Risk factors include the following: waist ≥88cm, high BP ≥90th percentile for age, sex and height, HDL <50 mg/dL, LDL-C ≥110mg/dL, TG ≥110mg/dL, HOMA-IR ≥4.
Mean intakes ages 9–17
Table 3 shows CMR clustering associated with three empirically-defined dietary patterns consisting of the following: (a) combined dairy and FnsV intakes, (b) combined dairy and grain intakes, and (c) combined FnsV and grain intakes. Higher combined intakes in each of these three healthy diet patterns were associated with less clustering of CMR. For example, only 9.3% girls with high FnsV and high grain intakes had three or more CMR factors compared with 21.1% girls with low intakes of both in early adolescence. Similarly, only 10.2% girls with higher intakes of both dairy and FnsV had three or more risk factors compared with 19.5% of those with low intakes. Diet patterns characterized by high intakes of FnsV alone or high dairy alone were somewhat less beneficial than the combined diet patterns.
Table 3.
Clustering of cardiometabolic risk factors at 18–20 years associated with dietary patterns
| Number of Risk Factorsa | ||||||
|---|---|---|---|---|---|---|
| Dietary Intakesb (ages 9–17 yrs) | N | 0 | 1 | 2 | 3 | 4–6 |
| N (row percent) | ||||||
| Dairy & FnsV | ||||||
| Low Da / Low FnsV | 635 | 177 (27.9%) | 216 (34.0%) | 118 (18.6%) | 70 (11.0%) | 54 (8.5%) |
| Low Da / High FnsV | 341 | 107 (31.4%) | 120 (35.2%) | 53 (15.5%) | 36 (10.6%) | 25 (7.3%) |
| High Da / Low FnsV | 206 | 60 (29.1%) | 72 (35.0%) | 26 (12.6%) | 33 (16.0%) | 15 (7.3%) |
| High Da / High FnsV | 187 | 68 (36.4%) | 70 (37.4%) | 30 (16.0%) | 13 (7.0%) | 6 (3.2%) |
| p-value for χ2 | 0.0046 | |||||
| Dairy & Grains | ||||||
| Low Da / Low Gr | 775 | 224 (28.9%) | 263 (33.9%) | 133 (17.2%) | 85 (11.0%) | 70 (9.0%) |
| Low Da / High Gr | 201 | 60 (29.9%) | 73 (36.3%) | 38 (18.9%) | 21 (10.5%) | 9 (4.5%) |
| High Da / Low Gr | 245 | 77 (31.4%) | 82 (33.5%) | 37 (15.1%) | 32 (13.1%) | 17 (6.9%) |
| High Da / High Gr | 148 | 51 (34.5%) | 60 (40.5%) | 19 (12.8%) | 14 (9.5%) | 4 (2.7%) |
| p-value for χ2 | 0.0087 | |||||
| FnsV & Grains | ||||||
| Low FnsV / Low Gr | 664 | 186 (28.0%) | 223 (33.6%) | 115 (17.3%) | 80 (12.1%) | 60 (9.0%) |
| Low FnsV / High Gr | 177 | 51 (28.8%) | 65 (36.7%) | 29 (16.4%) | 23 (13.0%) | 9 (5.1%) |
| High FnsV / Low Gr | 356 | 115 (32.3%) | 122 (34.3%) | 55 (15.5%) | 37 (10.4%) | 27 (7.6%) |
| High Fnsv / High Gr | 172 | 60 (34.9%) | 68 (39.5%) | 28 (16.3%) | 12 (7.0%) | 4 (2.3%) |
| p-value for χ2 | <0.001 | |||||
FnsV = fruit and non-starchy vegetables; BP = blood pressure; HDL = high-density lipoprotein: LDL = low-density lipoprotein; TG = triglycerides; HOMA-IR = homeostasis model assessment -insulin resistance
Risk factors include the following: waist ≥88cm, BP ≥90th percentile for age, sex and height, HDL <50 mg/dL, LDL-C ≥110mg/dL, TG ≥110mg/dL, HOMA-IR ≥4.
Low vs. high dairy=<2 vs. ≥2 cup-eq; low vs. high FnsV=<1.5 vs. ≥ 1.5 cup-eq; low vs. high grain=<7 vs. ≥7 oz-eq (based on mean dietary intakes ages 9–17).
Table 4 shows the effect of diet patterns on the risk of CMR clustering after adjusting for age, race, SES, height, physical activity, and hours spent watching television/video. Girls with any of the three healthy eating patterns shown here were about 40% less likely to have two or more CMR factors than girls with lower intakes in these food groups. They were even less likely to have three or more clustered risk factors. For example, girls with high intakes of dairy and FnsV (vs. those with lower intakes) were nearly 48% less likely (RR=0.52; 95% CI: 0.30, 0.89) to have three or more CMR factors.
Table 4.
Relative risk of two or more cardiometabolic risk factors at 18–20 years associated with dietary patterns
| # with >2 risk factorsb | # with >3 risk factorsb | |||
|---|---|---|---|---|
| Dietary Intakesa (ages 9–17 yrs) | RRc | 95% CI | RRc | 95% CI |
| Dairy & FnsV | ||||
| Low Da / Low FnsV | 1.00 | - | 1.00 | - |
| Low Da / High FnsV | 0.87 | 0.65–1.16 | 0.98 | 0.69–1.39 |
| High Da / Low FnsV | 0.91 | 0.64–1.28 | 1.28 | 0.85–1.91 |
| High Da / High FnsV | 0.62 | 0.42–0.91 | 0.52 | 0.30–0.89 |
| Dairy & Grains | ||||
| Low Da / Low Gr | 1.00 | - | 1.00 | - |
| Low Da / High Gr | 0.84 | 0.60–1.17 | 0.67 | 0.43–1.03 |
| High Da / Low Gr | 0.94 | 0.68–1.30 | 1.06 | 0.71–1.56 |
| High Da / High Gr | 0.57 | 0.37–0.86 | 0.56 | 0.33–0.97 |
| FnsV & Grains | ||||
| Low FnsV / Low Gr | 1.00 | - | 1.00 | - |
| Low FnsV / High Gr | 0.84 | 0.57–1.14 | 0.78 | 0.50–1.19 |
| High FnsV / Low Gr | 0.86 | 0.65–1.15 | 0.89 | 0.63–1.26 |
| High Fnsv / High Gr | 0.58 | 0.40–0.86 | 0.41 | 0.23–0.71 |
FnsV = fruit and non-starchy vegetables; Gr = grains; BP = blood pressure; HDL = high-density lipoprotein; LDL = low-density lipoprotein; TG = triglycerides; HOMA-IR = homeostasis model assessment-insulin resistance
Low vs. high dairy=<2 vs. ≥2 cup-eq; low vs. high FnsV=<1.5 vs. ≥ 1.5 cup-eq; low vs. high grain=<7 vs. ≥7 oz-eq (based on mean dietary intakes ages 9–17).
Risk factors include the following: waist ≥88cm, BP ≥90th percentile for age, sex and height, HDL <50 mg/dL, LDL-C ≥110mg/dL, TG ≥110mg/dL, HOMA-IR ≥4.
Adjusted for age (at time of diet assessment), race, SES, mean height, physical activity, and television/video watching (hours/day).
Discussion
These results suggest that earlier adolescent diet patterns are important predictors of subsequent clustering of cardiometabolic risk. Girls with higher usual intakes of total dairy, fruit and nonstarchy vegetables and grains had less accumulated risk by the time of late adolescence. Specifically, higher combined intakes of dairy and FnsV as well as higher intakes of FnsV and grains were associated with a 40–60% lower risk of CMR clustering.
Food-based dietary determinants of cardiometabolic risk clustering in younger populations have not been widely studied. Published data from the IDEFICS study of over 16,000 children ages 29 from eight European countries examined the association between individual food components and the clustering of cardiovascular risk.[8] Higher intakes of breakfast cereals as well as chocolate, nut-based spreads, and honey were positively associated with lower cardiovascular risk clustering, particularly among older subjects. Recently published cross-sectional data from another large European study of adolescents suggested that higher intakes of dairy were linked with less clustering of cardiovascular risk, particularly among female subjects.[23]
Few studies have used longitudinal data to examine the association between food intake patterns during the adolescent years and the clustering of CMR factors in late adolescence. Several studies have compared the association between dietary patterns derived from factor analyses, typically a “healthy” diet vs. a “less healthy” diet, and selected cardiovascular risk factors. Cross-sectional data from one Australian adolescent cohort found that a “Western” diet pattern was associated with a greater likelihood of metabolic risk clustering [24]. Since that study did not provide data on individual food groups, it is not possible to compare results directly. Analysis of 1999–2002 NHANES cross-sectional data suggests a lower prevalence of metabolic syndrome in adolescents with higher diet quality as measured by the Healthy Eating Index [25].
Several additional studies have examined the effects of selected food groups on risk of metabolic syndrome in adolescents and young adults. Results from the cross-sectional CASPIAN Study of school-aged children showed that increased consumption of dairy as well as fruits and vegetables was associated with a decreased risk of metabolic syndrome [26]. Another cross-sectional study found that markers of inflammation and oxidative stress were also inversely associated with fruit and vegetable intake among adolescents [27]. In the Bogalusa Heart Study, higher fruit and vegetable consumption was also linked with a lower prevalence of metabolic syndrome in adolescents and young adults (ages 19–38 years) [28]. In the CARDIA study of young adults, diet patterns that included higher dairy intakes led to a lower risk of insulin resistance syndrome [29].
A number of potential mechanisms could explain these results. First, fruits and vegetables are key sources of nutrients, fiber, and other phytochemicals and thus have the potential to lower oxidative stress, decrease platelet aggregation, and reduce cholesterol via lowered hepatic cholesterol synthesis and/or increased excretion of fecal fatty acids. Magnesium in particular, which is present in many fruits and vegetables as well as dairy, is thought to act as a natural calcium channel blocking agent and to improve vascular resistance [30], thereby benefiting blood pressure. Fruits and vegetables are also nutrient-rich, low energy-dense foods that may lead to reductions in overall energy intake, providing benefits to the regulation of body weight [31]. Fruits and vegetables are also a key source of fiber which is believed to benefit lipid profiles, particularly in terms of lowering LDL [32].
While dairy has been linked with lower levels of body fat in some studies of children and adolescents [33], others have found no effect. Two recent reviews found that while there is insufficient evidence to conclude that dairy has a beneficial effect on body weight/fat, there is substantial evidence that it has no adverse effect [34;35]. The protein content of dairy is one of several mechanisms by which it may convey benefits to body fat or composition [36]. In addition to its magnesium content, dairy is also an important source of both calcium and potassium, minerals that have been associated with lower blood pressures through the promotion of vasodilation and reduction of vascular resistance. Since a low potassium-to-sodium ratio decreases synthesis of nitric oxide and increases renal sodium retention, it may be responsible for elevations in blood pressure. Consequently, higher potassium intake may reduce salt sensitivity and lower blood pressure by reducing peripheral vascular resistance. Dairy-derived tripeptides are another potential pathway to improvements in blood pressure. A meta-analysis of randomized controlled trials demonstrated important hypotensive effects of such tripeptides [37], which lead to inhibition of the renin-angiotensin system, thereby lowering blood pressure.
Numerous studies address the metabolic health benefits of whole grains in adults. One such study amongst adolescents found beneficial effects of whole grain consumption on both body mass and insulin sensitivity, particularly among the heaviest subjects [38]. Few studies have examined the effects of total grains on cardiometabolic risk in children or adolescents. Data from the third National Health and Nutrition Examination Survey showed that total grain consumption was inversely associated with anthropometric measures of central adiposity [39]. While several studies of children and adolescents have found beneficial effects of overall grain consumption, most studies have attributed those benefits to whole grains [40]. A recently published review concluded that the vast majority of prospective studies of refined grain consumption (largely amongst adults) found either no effect or a protective effect of higher levels of intake on CVD risk [41]. Potential beneficial effects of refined grains may stem from its nutrient composition (e.g., B-vitamins) or from an association with other dietary habits (e.g., lower fat intakes).
The importance of cardiovascular risk clustering is well established and risk scores have been used to predict the development of coronary heart disease amongst adults for a number of years. Longitudinal data from the Bogalusa Heart Study further indicates an association between the number of cardiovascular risk factors in childhood and the likelihood of atherosclerotic lesions in early adulthood [42]. Data also suggests that a cluster score derived from the components of the metabolic syndrome in late childhood (waist circumference, triglycerides, HDL, fasting glucose, and blood pressure) was a better predictor of early adult cardiovascular risk than the simple identification of “metabolic syndrome” or any single component of the cluster [43]. A study examined cross-sectional data among 474 adolescents and found that those subjects who had ≥2 cardiovascular risk factors (vs. less) had greater vascular stiffness and higher wall thickness compared with those having few risk factors [3].
This study has several important strengths including its prospective design with ten years of follow up throughout adolescence, the large number of 3-day diet records which enhance the precision of estimated dietary intakes, and the replicate measures of cardiovascular risk factors and potential confounders. Calculation of food group intakes by linking NDS data from study subjects with USDA Food Pyramid that underlies the analysis is also an important strength. This study is, however, subject to several limitations that are common to epidemiologic studies of dietary exposures. Dietary data are of necessity obtained by self-report and are thus subject to potential misreporting such as underestimation of energy intake. A further specific limitation of the study is the low levels of intake in certain food sub-groups (e.g., whole grains) which limited the power of certain sub-analyses related to these exposures.
The results of this study support the hypothesis that food intake patterns in early to midadolescence may have long-term effects on the clustering of cardiometabolic risk in older African-American and Caucasian adolescent girls. These findings may have important implications for the prevention of cardiovascular disease and diabetes in these populations.
Acknowledgment:
This manuscript uses previously collected “NHLBI Growth and Health Study” research data obtained from the National Heart Lung and Blood Institute. The analyses were supported by Grant # R21DK075068 from the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) and a grant from the National Dairy Council.
Footnotes
Conflict of Interest: The authors declare no conflict of interest.
Reference List
- 1.Camhi SM, Katzmarzyk PT (2010) Tracking of cardiometabolic risk factor clustering from childhood to adulthood. Int J Pediatr Obes 5:122–129 [DOI] [PubMed] [Google Scholar]
- 2.Hakanen M, Lagstrom H, Pahkala K, Sillanmaki L, Saarinen M, Niinikoski H, Raitakari OT, Viikari J, Simell O, Ronnemaa T (2010) Dietary and lifestyle counselling reduces the clustering of overweight-related cardiometabolic risk factors in adolescents. Acta Paediatr 99:888–895 [DOI] [PubMed] [Google Scholar]
- 3.Shah AS, Dolan LM, Gao Z, Kimball TR, Urbina EM (2011) Clustering of risk factors: a simple method of detecting cardiovascular disease in youth. Pediatrics 127:e312–e318 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Djousse L, Padilla H, Nelson TL, Gaziano JM, Mukamal KJ (2010) Diet and metabolic syndrome. Endocr Metab Immune Disord Drug Targets 10:124–137 [DOI] [PubMed] [Google Scholar]
- 5.Berg CM, Lappas G, Strandhagen E, Wolk A, Toren K, Rosengren A, Aires N, Thelle DS, Lissner L (2008) Food patterns and cardiovascular disease risk factors: the Swedish INTERGENE research program. Am J Clin Nutr 88:289–297 [DOI] [PubMed] [Google Scholar]
- 6.Crichton GE, Bryan J, Buckley J, Murphy KJ (2011) Dairy consumption and metabolic syndrome: a systematic review of findings and methodological issues. Obes Rev 12:e190–e201 [DOI] [PubMed] [Google Scholar]
- 7.Feldeisen SE, Tucker KL (2007) Nutritional strategies in the prevention and treatment of metabolic syndrome. Appl Physiol Nutr Metab 32:46–60 [DOI] [PubMed] [Google Scholar]
- 8.Bel-Serrat S, Mouratidou T, Bornhorst C, Peplies J, De HS, Marild S, Molnar D, Siani A, Tornaritis M, Veidebaum T, Krogh V, Moreno LA (2012) Food consumption and cardiovascular risk factors in European children: the IDEFICS study. Pediatr Obes 8:225–236 [DOI] [PubMed] [Google Scholar]
- 9.Berz JP, Singer MR, Guo X, Daniels SR, Moore LL (2011) Use of a DASH Food Group Score to Predict Excess Weight Gain in Adolescent Girls in the National Growth and Health Study. Arch Pediatr Adolesc Med 165:540–546 [DOI] [PubMed] [Google Scholar]
- 10.Moore LL, Bradlee ML, Singer MR, Qureshi MM, Buendia JR, Daniels SR (2012) Dietary Approaches to Stop Hypertension (DASH) eating pattern and risk of elevated blood pressure in adolescent girls. Br J Nutr 108:1678–1685 [DOI] [PubMed] [Google Scholar]
- 11.Bradlee ML, Singer MR, Daniels SR, Moore LL (2013) Eating patterns and lipid levels in older adolescent girls. Nutr Metab Cardiovasc Dis 23:196–204 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Obarzanek E, Schreiber GB, Crawford PB, Goldman SR, Barrier PM, Frederick MM, Lakatos E (1994) Energy intake and physical activity in relation to indexes of body fat: the National Heart, Lung, and Blood Institute Growth and Health Study. Am J Clin Nutr 60:15–22 [DOI] [PubMed] [Google Scholar]
- 13.Schakel SF, Sievert YA, Buzzard IM (1988) Sources of data for developing and maintaining a nutrient database. J Am Diet Assoc 88:1268–1271 [PubMed] [Google Scholar]
- 14.Cook A, Friday J (2004) Pyramid Servings Database for USDA Survey Food Codes Version 2.0 Beltsville,MD: USDA, Agricultural Research Service, Community Nutrition Research Group; http://www.barc.usda.gov/bhnrc/cnrg. [Google Scholar]
- 15.Friedewald WT, Levy RI, Fredrickson DS (1972) Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem 18:499–502 [PubMed] [Google Scholar]
- 16.Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC (1985) Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 28:412–419 [DOI] [PubMed] [Google Scholar]
- 17.Kimm SY, Glynn NW, Kriska AM, Fitzgerald SL, Aaron DJ, Similo SL, McMahon RP, Barton BA (2000) Longitudinal changes in physical activity in a biracial cohort during adolescence. Med Sci Sports Exerc 32:1445–1454 [DOI] [PubMed] [Google Scholar]
- 18.National Cholesterol Education Program (2001) Executive Summary of The Third Report of The National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III). JAMA 285:2486–2497 [DOI] [PubMed] [Google Scholar]
- 19.Zimmet P, Alberti KG, Kaufman F, Tajima N, Silink M, Arslanian S, Wong G, Bennett P, Shaw J, Caprio S (2007) The metabolic syndrome in children and adolescents - an IDF consensus report. Pediatr Diabetes 8:299–306 [DOI] [PubMed] [Google Scholar]
- 20.National High Blood Pressure Education Program Working Group on Hypertension Control in Children and Adolescents (1996) Update on the 1987 Task Force Report on High Blood Pressure in Children and Adolescents: a working group report from the National High Blood Pressure Education Program. Pediatrics 98:649–658 [PubMed] [Google Scholar]
- 21.American Academy of Pediatrics (1992) National Cholesterol Education Program: Report of the Expert Panel on Blood Cholesterol Levels in Children and Adolescents. Pediatrics 89:525–584 [PubMed] [Google Scholar]
- 22.Ascaso JF, Romero P, Real JT, Lorente RI, Martinez-Valls J, Carmena R (2003) Abdominal obesity, insulin resistance, and metabolic syndrome in a southern European population. Eur J Intern Med 14:101–106 [DOI] [PubMed] [Google Scholar]
- 23.Bel-Serrat S, Mouratidou T, Jimenez-Pavon D, Huybrechts I, Cuenca-Garcia M, Mistura L, Gottrand F, Gonzalez-Gross M, Dallongeville J, Kafatos A, Manios Y, Stehle P, Kersting M, De HS, Castillo M, Hallstrom L, Molnar D, Widhalm K, Marcos A, Moreno LA (2014) Is dairy consumption associated with low cardiovascular disease risk in European adolescents? Results from the HELENA Study. Pediatr Obes 9:401–410 [DOI] [PubMed] [Google Scholar]
- 24.Ambrosini GL, Huang RC, Mori TA, Hands BP, O'Sullivan TA, de Klerk NH, Beilin LJ, Oddy WH (2010) Dietary patterns and markers for the metabolic syndrome in Australian adolescents. Nutr Metab Cardiovasc Dis 20:274–283 [DOI] [PubMed] [Google Scholar]
- 25.Pan Y, Pratt CA (2008) Metabolic syndrome and its association with diet and physical activity in US adolescents. J Am Diet Assoc 108:276–286 [DOI] [PubMed] [Google Scholar]
- 26.Kelishadi R, Gouya MM, Adeli K, Ardalan G, Gheiratmand R, Majdzadeh R, Mahmoud-Arabi MS, Delavari A, Riazi MM, Barekati H, Motaghian M, Shariatinejad K, Heshmat R (2008) Factors associated with the metabolic syndrome in a national sample of youths: CASPIAN Study. Nutr Metab Cardiovasc Dis 18:461–470 [DOI] [PubMed] [Google Scholar]
- 27.Holt EM, Steffen LM, Moran A, Basu S, Steinberger J, Ross JA, Hong CP, Sinaiko AR (2009) Fruit and vegetable consumption and its relation to markers of inflammation and oxidative stress in adolescents. J Am Diet Assoc 109:414–421 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Yoo S, Nicklas T, Baranowski T, Zakeri IF, Yang SJ, Srinivasan SR, Berenson GS (2004) Comparison of dietary intakes associated with metabolic syndrome risk factors in young adults: the Bogalusa Heart Study. Am J Clin Nutr 80:841–848 [DOI] [PubMed] [Google Scholar]
- 29.Pereira MA, Jacobs DR Jr., Van Horn L, Slattery ML, Kartashov AI, Ludwig DS (2002) Dairy consumption, obesity, and the insulin resistance syndrome in young adults:The CARDIA Study. JAMA 287:2081–2089 [DOI] [PubMed] [Google Scholar]
- 30.Houston MC, Harper KJ (2008) Potassium, magnesium, and calcium: their role in both the cause and treatment of hypertension. J Clin Hypertens 10:3–11 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Rolls BJ, Ello-Martin JA, Tohill BC (2004) What can intervention studies tell us about the relationship between fruit and vegetable consumption and weight management? Nutr Rev 62:1–17 [DOI] [PubMed] [Google Scholar]
- 32.Marlett JA, McBurney MI, Slavin JL (2002) Position of the American Dietetic Association: health implications of dietary fiber. J Am Diet Assoc 102:993–1000 [DOI] [PubMed] [Google Scholar]
- 33.Berkey CS, Rockett HR, Willett WC, Colditz GA (2005) Milk, dairy fat, dietary calcium, and weight gain: a longitudinal study of adolescents. Arch Pediatr Adolesc Med 159:543–550 [DOI] [PubMed] [Google Scholar]
- 34.Louie JC, Flood VM, Hector DJ, Rangan AM, Gill TP (2011) Dairy consumption and overweight and obesity: a systematic review of prospective cohort studies. Obes Rev 12:e582–e592 [DOI] [PubMed] [Google Scholar]
- 35.Spence LA, Cifelli CJ, Miller GD (2011) The role of dairy products in healthy weight and body composition in children and adolescents. Curr Nutr Food Sci 7:40–49 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Luhovyy BL, Akhavan T, Anderson GH (2007) Whey proteins in the regulation of food intake and satiety. J Am Coll Nutr 26:704S–712S [DOI] [PubMed] [Google Scholar]
- 37.Kris-Etherton PM, Grieger JA, Hilpert KF, West SG (2009) Milk products, dietary patterns and blood pressure management. J Am Coll Nutr 28 Suppl 1:103S–119S [DOI] [PubMed] [Google Scholar]
- 38.Steffen LM, Jacobs DR Jr., Murtaugh MA, Moran A, Steinberger J, Hong CP, Sinaiko AR (2003) Whole grain intake is associated with lower body mass and greater insulin sensitivity among adolescents. Am J Epidemiol 158:243–250 [DOI] [PubMed] [Google Scholar]
- 39.Bradlee ML, Singer MR, Qureshi MM, Moore LL (2010) Food group intake and central obesity among children and adolescents in the Third National Health and Nutrition Examination Survey (NHANES III). Public Health Nutr 13:797–805 [DOI] [PubMed] [Google Scholar]
- 40.Newby PK, Peterson KE, Berkey CS, Leppert J, Willett WC, Colditz GA (2003) Dietary composition and weight change among low-income preschool children. Arch Pediatr Adolesc Med 157:759–764 [DOI] [PubMed] [Google Scholar]
- 41.Williams PG (2012) Evaluation of the evidence between consumption of refined grains and health outcomes. Nutrition Reviews 70:80–90 [DOI] [PubMed] [Google Scholar]
- 42.Berenson GS, Srinivasan SR, Bao W, Newman WP III, Tracy RE, Wattigney WA (1998) Association between multiple cardiovascular risk factors and atherosclerosis in children and young adults. The Bogalusa Heart Study. N Engl J Med 338:1650–1656 [DOI] [PubMed] [Google Scholar]
- 43.Kelly AS, Steinberger J, Jacobs DR, Hong CP, Moran A, Sinaiko AR (2011) Predicting cardiovascular risk in young adulthood from the metabolic syndrome, its component risk factors, and a cluster score in childhood. Int J Pediatr Obes 6:e283–e289 [DOI] [PMC free article] [PubMed] [Google Scholar]