Nutrition Updates


  • KE D'Anci


Shellhaas RA, Barks AK, and Joshi SM. Prevalence and risk factors for vitamin D insufficiency among children with epilepsy. Pediatr Neurol. 2010;42:422–426.

The prevalence of vitamin D deficiency in children is increasing and may be related to decreased outdoor play and poor diet. Children with epilepsy may be at greater risk for vitamin D insufficiency due to the use of antiepileptic drugs, which are associated with low vitamin D status and bone fracture in adults. Older antiepileptic drugs induce hepatic cytochrome P450 metabolism, and it is posited that these antiepileptic drugs increase metabolism in the liver and lead to increased bone turnover. Newer antiepileptic drugs are less potent enzyme inducers, but they may still impact bone metabolism. Of concern in children is the use of antiepileptic drugs during periods of peak bone accrual, as this could affect skeletal health over the lifespan. The goal of the study by Shellhaas et al. was to assess the risk factors for and prevalence of vitamin D insufficiency in a pediatric population of epilepsy patients. (In a related paper in the same issue of the journal, the topic is reviewed in great detail by Shellhaas and Joshi.)

Children with epilepsy (n = 78) who were between the ages of 3 and 17 years old were enrolled in the study from the authors' private practice at the University of Michigan Medical Center. Children were classified according to epilepsy type, seizure frequency, type of pharmacotherapy, and duration of therapy. Additionally, ambulatory status, mental status, body mass index, sex, and ethnicity were included as cofactors. Serum measurements of 25-hydroxyvitamin D and 1,25-dihydroxyvitamin D were collected in the fall and winter months to reduce variability from recent sun exposure. None of the patients had abnormal levels of 1,25-dihydroxyvitamin D, but approximately 75% of the children had 25-hydroxyvitamin D levels that were below normal (<32 ng/mL). The mean 25-hydroxyvitamin D level was 28.37 ng/mL, and 25% of the children had a 25-hydroxyvitamin D level below 20 ng/mL. Logistic regression analysis with a dichotomous outcome of less than or greater than 25 ng/mL 25-hydroxyvitamin D for low versus normal was used, although current research suggests that 32 ng/mL should be considered the cutoff. The authors indicate that so few children had 25-hydroxyvitamin D values of 32 ng/mL that there was insufficient power to conduct an analysis. Using this model, female gender and higher BMI as well as localized (as opposed to generalized) epilepsy were significantly predictive of vitamin D insufficiency. However, children with localized epilepsy had higher BMIs and the authors acknowledge this may be a confounder in the association. No differences were seen regarding the risk of vitamin D insufficiency in relation to the type of treatment, cerebal palsy, seizure control, and mental ability; the authors indicate this observation is important since clinicians may not be able to reliably predict which patients have low vitamin D levels. In the present study, bone mineral density was not measured, but the association between low vitamin D status and abnormal bone metabolism suggest that pediatric epileptic patients may be at risk for poor bone health.

Comment: In accompanying commentary, Dr. Wirrell raises questions about the relationship between vitamin D insufficiency in children with epilepsy and potential poor bone health. To what degree can improving the diet and encouraging physical activity alleviate the negative impact of epilepsy on vitamin D metabolism and bone health? Should bone-mineral density scans be performed in children with a history of fracture? With respect to supplementation, she concurs that all children with epilepsy be given a minimum of 400 IU/day of vitamin D (which is the RDA), and she argues that up to 1,000 IU/day could be of benefit for children with symptomatic epilepsy, those with impaired mobility, and those treated with more than one antiepileptic medication.

Comment: Wirrell E. Vitamin D and bone health in children with epilepsy: Fad or fact? Pediatr Neurol. 2010;42:394–395.


Maffeis C, Banzato C, Brambilla P, Cerutti F, Corciulo N, Cuccarolo G, Di Pietro M, Franzese A, Gennari M, Balsamo A, Grugni G, Iughetti L, Del Giudice EM, Petri A, Trada M, Yiannakou P; and the Obesity Study Group of the Italian Society of Pediatric Endocrinology and Diabetology. Insulin resistance is a risk factor for high blood pressure regardless of body size and fat distribution in obese children. Nutr Metab Cardiovasc Dis. 2010;20:266–273.

Hypertension in children has historically been rare, but recent work shows an increasing prevalence of high blood pressure in children. As in adults, hypertension in children is associated with increased cardiovascular disease risk factors, thickening of arterial walls, and overall morbidity. It is suggested that the rapid increase in high blood pressure in children is related to the recent increased prevalence of obesity. Some research shows a relationship between fasting insulin levels and blood pressure in obese children, but this relationship is not well understood. The present study by Maffeis et al. assessed the relationship between high blood pressure, insulin resistance, and body fat distribution in overweight and obese children.

Overweight and obese children (N = 1,044; 560 girls) between the ages of 6 and 11 years were recruited from obesity clinical care centers in Italy. Fasting insulin, glucose, and blood lipids measurements were collected. Body weight, height, and waist circumference were measured and used to generate BMI and abdominal adiposity scores. Measurements for hypertension and adiposity were calculated according to age-appropriate modifications. Children were classified as normotensive if both systolic and diastolic measures were <95th percentile for age, sex, and height. Children at or above the 95th percentile were classified as hypertensive. BMIs were compared to growth charts; children with BMIs above the curve for a BMI of 25 at age 18 years were considered overweight, and those falling above the curve for a BMI of 30 at age 18 years were considered obese. Fasting measures of insulin and blood glucose were used to calculate insulin resistance (HOMAIR). Children were categorized as prepubertal or pubertal based on Tanner stages, which use visual identification of primary and secondary indices of sexual development. The prevalence of hypertension was higher in males than in females for both overweight and obese children. Hypertension increased in direct association with increased BMI and was not related to pubertal stage. Of the biochemical measurements, only HOMAIR was predictive of hypertension, and the highest HOMAIR category regardless of weight status and fat distribution. The authors suggest that, in overweight and obese children, measures of insulin resistance could be useful for identifying children at high risk for hypertension.


Venezia A, Barba G, Russo O, Capasso C, De Luca V, Farinaro E, Cappuccio FP, Galletti F, Rossi G, and Strazzullo P. Dietary sodium intake in a sample of adult male population in southern Italy: results of the Olivetti Heart Study. Eur J Clin Nutr. 2010;64:518–524.

Excess sodium in the diet is linked with increased blood pressure and increases risk for heart attack and stroke. Over the past 40 years, government agencies and health-professional organizations have engaged in global efforts to reduce consumption of excess sodium, but these efforts have been largely unsuccessful. In Italy, as in other Western countries, sodium intake remains high, although relatively few studies have examined the Italian population. In the present report by Venezia et al., dietary intakes of sodium were assessed in adult men (age range: 25–75 years) living in southern Italy.

Participants (N = 940) in the analysis represent a subset of individuals recruited into the Olivetti Heart Study, which is an epidemiological study of cardiovascular risk factors that began in 1975. Diet, anthropomorphic indices, and hypertension measurements were obtained during the 2002–2004 follow-up examination. Dietary sodium intake was assessed using food frequency questionnaires. Foods and food groups chosen for the analysis included sodium-rich foods or foods to which discretionary salt is typically added (e.g. bread, pasta, cheeses, cold cuts, and meats). 24-h urine samples were also collected, and urinary sodium excretion was used to estimate daily sodium intake. Blood pressure, blood lipids, blood glucose, body weight, height, and waist circumference were measured. The average sodium intake, estimated using 24-h urinary excretion, for adult southern Italian men was roughly 4,669 mg/g (or 12 g NaCl/day). Intake of cold cuts and pasta were the best dietary predictors of high sodium intake, and higher BMI was the best overall predictor of higher sodium intake. Higher salt intake and lower sodium excretion was seen in overweight and obese participants. There were no observed differences in sodium intake or excretion in individuals with hypertension in comparison to normotensive participants (50th percentile: hypertensive = 4,437 mg/day; normotensive = 4,761 mg/day). When comparing treated versus untreated hypertensive individuals, sodium intakes tended to be lower (50th percentile: untreated = 4,623 mg/day; treated = 4,328 mg/day). Sodium intakes in the present analysis were similar to and, indeed, higher than the 4,140 mg/day intakes reported previously in a 1976 subgroup of the Olivetti Heart Study, and 94% of the study population ingested more than the 2,300 mg/day recommended by the Italian government; this, suggests that campaigns to reduce sodium intake have not benefitted this population. In overweight individuals, there was evidence that sodium handling was impaired relative to normal-weight individuals, and the authors suggested that overweight individuals would benefit to a greater degree from reductions in sodium intake relative to normal-weight individuals.


Jakobsen MU, Dethlefsen C, Joensen AM, Stegger J, Tjønneland A, Schmidt EB, and Overvad K. Intake of carbohydrates compared with intake of saturated fatty acids and risk of myocardial infarction: importance of the glycemic index. Am J Clin Nutr. 2010;91:1764–1768.

Reducing intake of saturated fatty acids in the diet reduces the risk for cardiovascular disease and the risk for myocardial infarction. To assist individuals in their efforts to reduce intake of saturated fats, typical recommendations include reducing total fat intake while increasing intake of complex carbohydrates or to specifically reduce saturated fatty acid intake and increase polyunsaturated fatty acid intake. Epidemiological studies suggest that the risk for ischemic heart disease is increased when saturated fatty acids are replaced with carbohydrates and that the risk is lower when saturated fatty acids are replaced with polyunsaturated fatty acids. More recent work shows that the glycemic index of carbohydrates may be an important consideration, with intake of low-glycemic-index foods reducing heart attack risk. It is thought that dyslipidemia provoked by high consumption of high-glycemic index foods may increase the risk for ischemic heart disease. The present prospective cohort study by Jakobsen et al. examined the relative risk for myocardial infarction with lower energy intake from dietary saturated fatty acids and higher energy intake from carbohydrates.

The cohort of 53,644 individuals included in the study was a cancer-free subset of participants between the ages of 50 and 64 years who were enrolled into the Danish Diet, Cancer, and Health study between 1993 and 1997. At recruitment, food frequency and lifestyle questionnaires assessed nutrient intake, smoking habits, and physical activity levels. For each participant, the average dietary glycemic index value was calculated and diets were separated into glycemic index groups categorized as low (77–85), medium (86–90), or high (91–98). The primary outcome measure was incident fatal and nonfatal myocardial infarction (MI). Information on the incidence of MI was obtained from the Danish National Patient Registry, and incidence was tracked from enrollment until first MI, death (due to MI or any cause), emigration, or the year 2008, whichever came first. Estimated hazard ratios were interpreted as estimated differences in risk for a 5% higher energy intake from carbohydrates and a concomitant lower intake of saturated fatty acids (referred to as substitution). Over the median 12 years of follow-up, 1,943 cases of MI occurred. There was a significant positive association between substitution of high-glycemic-index carbohydrates for saturated fatty acids and risk for MI and a non-significant inverse association between substitution of low-glycemic-index carbohydrates for saturated fatty acids and risk for MI. No relationship was seen between substitution with medium-glycemic-index carbohydrates and risk for MI. In interpreting this study it should be remembered that the concept of “substitution” reported here represents an analysis of food frequency data rather than a dietary intervention. In the authors' analysis, the data suggest that individuals consuming less energy from saturated fatty acids and more energy from high-glycemic-index carbohydrates are at increased risk for MI in comparison with individuals who substitute low-glycemic-index carbohydrates. The authors suggest it is important to consider glycemic index when increasing energy from carbohydrates while reducing intake from saturated fatty acids.

Comment: In an accompanying editorial, Dr. Hu agrees that diets high in saturated fats or high in refined carbohydrates have little place in preventing ischemic heart disease. In light of the present study, and in context with existing data, he argues that a tight focus on reducing total and saturated fats may have limited utility as individuals typically replace fats with carbohydrates. He further proposes that the goal of dietary modification of heart disease risk should shift away from restricting fat intake and address instead reductions in refined carbohydrate intake.

Comment: Hu FB. Are refined carbohydrates worse than saturated fat? Am J Clin Nutr. 2010;91:1541–1542.