Low bone turnover and low bone density in a cohort of adults with Down syndrome.

UNLABELLED: Increased incidence of osteoporosis in Down syndrome has been reported, but etiology is not established. We report low bone turnover markers and bone mineral density (BMD) in a cohort of people with Down syndrome without consistent clinical risk factors. Our results should guide future studies and treatments for this common problem. INTRODUCTION: To better understand the etiology for osteoporosis in Down syndrome (DS), we measured bone density by dual-energy X-ray absorptiometry (DXA) and circulating biochemical markers of bone formation and resorption in a cohort of 30 community-dwelling DS adults. METHODS: Seventeen males and 13 females followed in the University of Arkansas Down Syndrome Clinic were evaluated by DXA to estimate BMD and underwent phlebotomy to measure serum procollagen type-1 intact N-terminal propeptide (P1NP) to evaluate bone formation, and serum C-terminal peptide of type-I collagen (CTx) to evaluate bone resorption. RESULTS: Seven of 13 DS females and 12 of 17 DS males had low bone mass at one of measured sites (z≤-2.0). When data were grouped by age, males had apparent osteopenia earlier than females. The mean P1NP in the normal group was 19.2±5.2 ng/ml vs. 2.2±0.9 ng/ml in the DS group (P=0.002). Serum CTx levels in the normal group were 0.4±0.1 ng/ml vs. 0.3±0.1 ng/ml (P=0.369). CONCLUSIONS: Low BMD in adults with DS is correlated with a significant decrease in bone formation markers, compared to controls without DS, and is independent of gender. These data suggest that diminished osteoblastic bone formation and inadequate accrual of bone mass, with no significant differences in bone resorption, are responsible for the low bone mass in DS. These observations question the use of antiresorptive therapy in this population and focus attention on increasing bone mass by other interventions.

Sources of variation in energy intake by men and women as determined from one year's daily dietary records.

Data from a year-long study conducted by the US Department of Agriculture's Beltsville Human Nutrition Research Center were used to identify sources of variation in daily energy intakes. A specific hypothesis was that an inverse relationship exists between past body weight and future energy intake whereas a direct relationship exists between past energy intake and future weight. Daily energy intakes of 29 male (n = 13) and female (n = 16) adult subjects were related through linear-regression analysis to sex (+ for males), age (-), height (+), weight (- for past and present measurements, + for future measurement), data collection method (- for duplicate plate), day of week (+ for Friday and Saturday), and month (+ for May, July, August, December, and January). These relationships were statistically significant (p less than 0.05). The results confirmed the hypothesis and were in general agreement with previously published studies that used larger samples but fewer days of data.

Mineral balances of men fed a diet containing fiber in fruits and vegetables and oxalic acid in spinach for six weeks.

In an investigation of the effects of fiber and oxalic acid on weekly mineral balances, 12 men consumed two diets consisting of natural foods for 6 wk each in a crossover design. One diet contained about 25 g neutral detergent fiber (NDF) in fruits and vegetables and included 100 g spinach, which is high in oxalic acid, every other day. The second diet was a low fiber diet that contained about 5 g NDF and the same amount of spinach as the first diet. On the basis of mean values for 6 wk, balances for calcium, magnesium, iron, zinc or manganese were not significantly different due to diet. Copper balances were significantly lower when the low fiber diet was consumed than when the diet containing fiber in fruits and vegetables and consequently a higher level of copper was consumed. Mineral balances were significantly different due to week. When results of both diets were considered together, mean mineral balances decreased at some time after wk 1 and returned to the wk 1 level during wk 6. Balances for wk 1 and 6 were not significantly different for any of the six minerals. These results demonstrate the importance of determining weekly balances for several weeks when conducting a human metabolic study.

Impact of variation in carbohydrate intake on mineral utilization by vegetarians.

Dietary intakes of two groups of vegetarians (of Asian Indian and of American origin) residing in the United States were compared with intakes of a group of nonvegetarians. Mineral intakes and balances were determined for all three groups for a 7-d period. The vegetarians had a significantly higher percent of energy intake from carbohydrate and significantly higher intakes of crude fiber (CF) and neutral detergent fiber (NDF) than did the nonvegetarians. The American vegetarians had significantly higher intakes of magnesium, iron, and copper than did the other groups. Manganese intakes were significantly higher for the two vegetarian groups than for the nonvegetarians. Calcium, iron, zinc, and copper balances were not significantly different among the groups. Magnesium and manganese balances were significantly more negative for the American vegetarians than for the other two groups. In general, the higher percent of carbohydrate intake or the higher level of fiber intake did not appear to affect mineral utilization by the vegetarians.

Effect of dietary fiber on the metabolizable energy of human diets.

Two diets consisting of natural foods that differed in fiber content were fed to 12 men for 6 wk each in a crossover design. Diet 1 contained 37 g total dietary fiber (TDF) from fruits and vegetables and diet 2 contained 16 g TDF from fruit and vegetable juices in a 2800-kcal menu. Seven-day food, urine and fecal composites were made and analyzed for their gross energy, fat, protein and carbohydrate content during the fourth and fifth weeks of each dietary consumption period. The metabolizable energy (ME) of these diets was determined (gross energy of the diet - gross energy of urine and feces) and calculated by Atwater's general energy conversion factors (4, 9, 4), by specific energy conversion factors used in a U.S. food table and by formulas derived by various researchers. The coefficients of availability of energy and the energy-containing nutrients, fat, protein and carbohydrate, were lower while subjects were consuming the higher fiber diet. The specific energy conversion factors of the U.S. food table gave better estimates of the true ME of the lower fiber diet than the higher fiber diet, overestimating the ME of these diets by only 6% (diet 1) and 4.6% (diet 2).

A study on Asian Indian and American vegetarians: indications of a racial predisposition to glucose intolerance.

Sixty-two Asian Indian and American vegetarians participated in a 3-h glucose tolerance test after an overnight fast to study clinical indices of glucose homeostasis. The Asian Indians had a higher (p less than 0.0035) insulinogenic score than the Americans. The Asian Indians had significantly higher insulin levels than the Americans at every sampling time during the glucose tolerance test except for the 3-h sample. The Indian men had significantly higher (p less than 0.05) plasma glucose than the other three groups at 2 h after the glucose load. American subjects had higher (p less than 0.0008) insulin binding to erythrocytes than the Asian Indian subjects. Scatchard analysis and competition-inhibition plots of the insulin-receptor data indicated that decreased binding in the Indian group results from a lowered number and decreased affinity of erythrocyte receptors. These results suggest that Asian Indians exhibit several clinical indications associated with an increased risk for the development of insulin-independent diabetes.

Effects of fiber, phytic acid, and oxalic acid in the diet on mineral bioavailability.

Studies on effects of fiber, phytic acid, and oxalic acid on mineral bioavailability are reviewed. It is difficult to separate the effects of fiber and phytic acid when cereal products are fed, because they occur together in these foods. However, even the combination of fiber and phytic acid does not appear to affect mineral balances unless unrefined cereal intakes are high and mineral intakes are low. Also, the body may be able to adjust to the decreased availability by increased absorption of the available mineral. Oxalic acid may result in decreases in mineral bioavailability if consumed with a high fiber diet, but the decrease may also be transient.

Number of days of food intake records required to estimate individual and group nutrient intakes with defined confidence.

The number of days of food intake data needed to estimate the intake of 29 male (n = 13) and female (n = 16) adult subjects, individually and as a group, was determined for food energy and 18 nutrients. The food intake records were collected in a year-long study conducted by the U.S. Department of Agriculture's Beltsville Human Nutrition Research Center. Each individual's average intake of nutrients and standard deviation over the year were assumed to reflect his or her "usual" intake and day-to-day variability. Confidence intervals (P less than 0.05) for each individual's usual intake were constructed, and from these the number of days of dietary records needed for estimated individual and group intake to be within 10% of usual intake was calculated. The results indicated that the number of days of food intake records needed to predict the usual nutrient intake of an individual varied substantially among individuals for the same nutrient and within individuals for different nutrients; e.g., food energy required the fewest days (averaging 31) and vitamin A the most (averaging 433). This was considerably higher than the number of days needed to estimate mean nutrient intake for this group, which ranged from 3 for food energy to 41 for vitamin A. Fewer days would be needed for larger groups.

Phytate:zinc and phytate X calcium:zinc millimolar ratios in self-selected diets of Americans, Asian Indians, and Nepalese.

The literature indicates that phytate:zinc and phytate X calcium:zinc millimolar ratios of a diet may provide useful indexes of the bioavailability of dietary zinc. However, there is little information on the phytate:zinc and phytate X calcium:zinc millimolar ratios for total human diets. We have therefore determined the phytate:zinc and phytate X calcium:zinc millimolar ratios of self-selected diets of 29 American omnivores, 23 American vegetarians, 30 Asian Indian immigrant vegetarians, and 26 Nepalese lactating vegetarians. Criteria for selection of subjects were: good health, no extreme dietary habits, and no intake of nutrient supplements. According to the limited literature, the suggested critical values for phytate:zinc and phytate X calcium:zinc millimolar ratios in animal diets and retrospective calculations from human diets are greater than 10 and greater than 200, respectively. The mean phytate:zinc and phytate X calcium:zinc millimolar ratios of American omnivorous diets in the present study were less than 10 and 200, respectively. Confirmation of the critical molar ratios as indexes of the bioavailability of zinc in human diets has not been established by experimentation. However, if the data from animal studies are applicable to human diets, the present study suggests that phytate has little influence on zinc bioavailability of most American diets. In contrast, the mean phytate:zinc and phytate X calcium:zinc millimolar ratios of all vegetarian diets were above the proposed critical levels. Those data, therefore, suggest that phytate might increase the risk of impaired zinc bioavailability for vegetarians consuming a relatively high level of calcium.

Carbohydrate tolerance and serum lipid responses to type of dietary carbohydrate and oral contraceptive use in young women.

The responses of serum glucose, insulin, and lipids to a glucose load were compared in two groups of college-age women fed two different carbohydrate diets. One group of women took oral contraceptives (OC) and the other group had never taken OC. Two experimental diets supplied approximately 13% of the calories from protein, 36% from fat, and 51% from carbohydrate. Of the carbohydrate, 84% was either sucrose or wheat starch. Each diet was fed for 4 weeks in a cross-over design. The subjects were given a glucose load dose, 1 g/kg of body weight, before and after each dietary period. Serum glucose, insulin, and lipids were measured before and 30, 60, 120, and 180 minutes after the load dose. Serum insulin was significantly affected by the type of carbohydrate in the diet (p less than .01). The sucrose diet generally gave lower insulin levels in response to the glucose load. Serum triglyceride levels were significantly higher (p less than .01) in users than nonusers of OC. Serum free fatty acids and total lipids were not significantly affected by OC use. There was no effect of different carbohydrates on any of the serum lipid parameters. Significant responses from fasting levels were observed for serum glucose, insulin, and triglyceride levels (p less than .01).

Fiber intakes, stool frequency, and stool weights of subjects consuming self-selected diets.

Neutral detergent fiber (NDF) contents of duplicate collections of food consumed by 29 men and women were determined for 7 days four times during the year. The ratio of analyzed NDF to calculated crude fiber intakes was approximately 2. Although mean NDF intake was 7.7 g/day for females and 9.5 g/day for males, the majority of the subjects had NDF intakes of from 4 to 9 g/day. Fiber-containing foods consumed with the greatest frequency were vegetables. NDF intakes were greater in the winter than in the spring and summer. Caloric and total carbohydrate intakes, number of stools, and fecal weights were greater for the males than for the females. Fecal weights were significantly correlated with caloric intake, NDF intake, and number of stools. It is suggested that the insoluble fiber measured by the NDF method is likely to be better correlated with fecal weight than the total dietary fiber which includes soluble fibers.

Seasonal variation in plasma glucose and hormone levels in adult men and women.

Twenty-nine men and women collected diet records for 1 yr. Fasting blood samples were collected four times during the year so that each season of the year was represented. An oral glucose tolerance test was given in the spring and in the fall. Regardless of season men consumed significantly more kilocalories, carbohydrate, sugar, and starch than did the women. Men consumed more simple sugars in the fall than in the spring. No seasonal effect was observed for calories, total carbohydrate, or starch intake. Fasting glucose and insulin were higher in the fall than in the spring. Summed insulin levels after the oral glucose tolerance were significantly different by age and season. Summed insulin levels were significantly higher in older than younger subjects; men and older subjects had higher levels in the spring than in the fall. Fasting levels of thyroxine, free thyroxine, triiodothyronine, and glucagon varied significantly with the season. A significant sex by season interaction was observed in the thyroid hormones. It appears that seasonal variation as well as diurnal variation should be considered in evaluating hormone levels.

Evaluation of long-term dietary intakes of adults consuming self-selected diets.

Mean daily intakes of calories and 19 nutrients were calculated for 13 males and 16 females, ages 20 to 53 yr, who kept daily diet records for 1 yr. Mean daily caloric and 19 selected nutrient intakes of the subjects met or exceeded the 1980 recommended dietary allowances except for iron and calcium for females. Males had higher intakes than females for all nutrients studied except crude fiber, vitamin A, and vitamin C. However, nutrient density values were comparable for both sexes, except that the females had higher nutrient density values for vitamin A than did the males. The younger subjects had higher intakes of calories and saturated fat than the older ones. The younger males had higher intakes of total fat, saturated fat, and oleic acid than the older males. The consistency of reporting food intakes examined by applying a systematic sampling method designed for this study did not vary considerably when diet records kept over a long period of time were evaluated by four methods.

Effect of making duplicate food collections on nutrient intakes calculated from diet records.

In a 1-yr study in which food intakes were recorded daily, subjects were asked to make duplicate food collections for 1 wk during each of the four seasons. Mean calculated energy intake of the 29 subjects was 12.9% less during the food collection periods than the mean for the entire year (which included collection periods). There were also significant reductions in the reported intake of all nutrients during the collection periods. Protein, vitamin A, saturated fat, and cholesterol intakes were decreased to the greatest extent. The actual decrease in energy intake was greater for the males than for the females, but the percentage was the same (12.9%). The younger group of subjects (less than or equal to 35) decreased energy intake 16.8% and the older group (greater than 35) 8.8%. Comparison of intakes during collection periods with intakes the week before and the week after showed that 28 of the 29 subjects decreased their energy intake from 1.1 to 32.3%. These data suggest that intakes of subjects during food collection periods do not represent their habitual levels of intake reported throughout the year.

Relationships among dietary constituents and specific serum clinical components of subjects eating self-selected diets.

Sex- and age-related differences in dietary and blood chemistry factors were investigated in subjects adhering to their usual lifestyles. Diet records were examined daily and blood chemistry profiles were monitored five times during the 1-yr study. As expected, high-density lipoprotein cholesterol was significantly higher in women than in men. Values of creatine phosphokinase, aspartic aminotransferase, alanine aminotransferase, glucose, triglycerides, urea nitrogen, uric acid, and total bilirubin were higher in men than in women. Glucose was lower while high-density lipoprotein cholesterol, albumin, and total protein were higher in the younger women than in older women. Alcohol consumption by men correlated positively with aspartic aminotransferase and alanine aminotransferase but not with high-density lipoprotein cholesterol. Alcohol consumption by women did correlate positively with high-density lipoprotein cholesterol but not with the aminotransferase enzymes. Correlations between serum high-density lipoprotein cholesterol and vitamin C intake were positive and significant in women. In men, high levels of high-density lipoprotein cholesterol seems to be associated with very high vitamin C intakes, but no associations were apparent at normal levels of these parameters. Serum cholesterol did not correlate significantly with dietary cholesterol, saturated fat, linoleic acid, or P/S in men or women.

Variations in plasma fatty acid concentrations during a one-year self-selected dietary intake study.

A group of healthy volunteers, maintaining their usual lifestyle, was monitored as to their nutrient intake for a period of 1 yr. Diet records were kept daily and blood samples were collected at even intervals five times during the year. Plasma fatty acid levels were analyzed to determine any sex, age, or seasonal variations or if the plasma fatty acid levels could be correlated to dietary fat intake. In the population studied, there was a significant (p less than 0.0001) sex by age interaction, but no seasonal effect was observed. No major differences in plasma fatty acids related to diet were found. However, the younger men had the highest linoleic acid intake and the lowest plasma linoleic acid. Plasma linoleic acid levels for males older than 35 yr of age (87.0 +/- 3.1 mg/dl) were significantly greater than for males younger than 36 yr of age (67.9 +/- 1.8 mg/dl). The plasma linoleic acid levels of all the women were intermediate in value to the men but not different from each other (73.8 +/- 1.7 mg/dl for the younger women and 78.7 +/- 1.9 mg/dl for the older women). There was a significant (p less than 0.01) sex effect in the relative percentage of plasma linoleic acid (34.4 +/- 0.4% for the women and 32.4 +/- 0.6% for the men), but no age or seasonal effect was observed.

Magnesium intakes, balances, and blood levels of adults consuming self-selected diets.

Magnesium balances and plasma and blood cell concentrations were determined in 34 men and women, 20 to 53 yr of age, during each season for 1 yr while they were consuming their customary diets. Dietary records were maintained 365 days. Analysis of variance indicated no seasonal variation in intake or balance. Blood cell magnesium correlated with magnesium intake (r2 = 0.203; p less than 0.04). Daily magnesium intakes of the men (323 mg) were significantly higher (p less than 0.01) than of the women (234 mg), and reflected greater calorie consumption for their greater body weights. Daily magnesium balances for the men and women were -32 and -25 mg, respectively. Multiple regression analysis showed that magnesium balance was associated not only with magnesium intake but also with the intake of other nutrients such as fiber, protein, calcium, and phosphorus. Moreover, the effect of these nutrients on magnesium balance varied with age and sex.

Zinc, copper, and manganese intake and balance for adults consuming self-selected diets.

Twenty-eight adult men and women participated in a year-long study designed to determine accurately the nutrient intake of adults who lived at home and consumed self-selected diets. During four metabolic balance periods, 7 days each, corresponding to the seasons (spring, summer, fall, winter), duplicates of the diet, and all urine and feces were collected. Daily mean intakes for zinc and copper were 9.9 and 1.2 mg, respectively. These levels were less than the recommended daily intakes of 15 mg for zinc and 2 to 3 mg for copper. In contrast, the mean dietary intake of manganese was 3.0 mg/day which is within the suggested safe and adequate range of 2.5 to 5.0 mg. Metabolic balances were negative for all three elements possibly due to a reduction in food intake during the collection periods compared to the noncollection intervals. The collection of the duplicate diets apparently influenced the food intake during the collection weeks.

Sodium and potassium intake and balance in adults consuming self-selected diets.

Twenty eight adults, 12 men and 16 women, participated in a 1-yr study designed to assess daily nutrient intake accurately. All subjects lived at home, consumed self-chosen diets, and maintained a detailed daily dietary record throughout the year. During four 7-day balance studies, one in each season of the year, meals, beverages, urine, and feces were analyzed for sodium and potassium content by atomic absorption spectrometry. Total intakes averaged 3.4 g/day for sodium and 2.8 g/day for potassium. The Na:K ratio for all diets analyzed averaged 1.3. Nutrient densities of sodium and potassium were 1.8 and 1.5 g/1000 kcal, respectively. Apparent absorptions of sodium and potassium were 98 and 85%, respectively, and did not change significantly over the wide range of intakes. Average urinary excretions of sodium and potassium were 86 and 77% of total intake, respectively. Mean metabolic balances were positive for sodium, +0.47 g/day, and potassium, +0.28 g/day. The data of this study provide useful information concerning the dietary intakes, excretions, and balances of sodium and potassium for adults based on analytic determination.