Hypovitaminosis D in obese children and adolescents: relationship with adiposity, insulin sensitivity, ethnicity, and season

Low 25-hydroxyvitamin D (25[OH] D) results in hyperparathyroidism and is among the endocrine derangements of adult obesity. There are differing recommendations on defining low 25(OH) D: hypovitaminosis D (serum 25[OH] D concentration <75 nmol/L) and vitamin D deficiency (serum 25[OH] D concentration <50 nmol/L). We sought to evaluate the prevalence of low levels of 25(OH) D by examining hypovitaminosis D (<75 nmol/L), vitamin D sufficiency (>= 75 nmol/L), vitamin D insufficiency (50-74.9 nmol/L), and vitamin D deficiency (<50 nmol/L) in pediatric obesity and the relationship to other calciotropic hormones and adiposity. Serum 25(OH) D, intact parathyroid hormone (iPTH), ionized calcium, glucose, and insulin levels along with hemoglobin A(1c) (HbA(1c)) and quantitative insulin sensitivity check index (QUICKI) were determined in 127 subjects aged 13.0 +/- 3.0 years (49 Caucasian [C], 39 Hispanic [H], and 39 African American [AA]; 61.2% female; body mass index 36.4 +/- 8.1 kg/m(2)) during fall/winter (F/W) and spring/summer (S/S). Body composition was determined by bioelectrical impedance. Hypovitaminosis D was present in 74% of the cohort, but was more prevalent in the H (76.9%, P < .05) and AA (87.2%, P < .05) groups than in the C group (59.1%). Hypovitaminosis D corresponded to decreased vitamin D intake (P < .005) and was more prevalent in F/W than S/S (98.4% vs 49.2, P < .01). Vitamin D deficiency was identified in 32.3% of the entire cohort and was more prevalent in the H (43.6%, P < .0001) and AA (48.7%, P < .0001) groups than in the C group (10.2%) associated with decreased vitamin D intake (P < .0001). Vitamin D insufficiency was present in 41.7% of the cohort, with similar prevalence among C (48.9%), H (33.3%), and AA (38.5%). Vitamin D insufficiency corresponded to decreased vitamin D intake (P < .005), with similar prevalence in F/W and S/S (45.3% vs 38.1%), whereas vitamin D deficiency was not only accompanied by decreased vitamin D intake (P < .0001) but was more prevalent in F/W than S/S (53.1% vs 11.1%, P < .0001). Serum 25(OH) D and iPTH (r = -0.41, P < .0001) levels were negatively correlated without seasonal and ethnic/racial influences. Hypovitaminosis D and vitamin D-deficient groups had higher body mass index, fat mass (FM), and iPTH, but had lower QUICKI than vitamin D-sufficient group (P < .01). Whereas FM was negatively correlated with 25(OH) D (r = -0.40, P < .0001), it was positively correlated with iPTH (r = 0.46, P <.0001) without seasonal and racial/ethnic influences. Serum 25(OH) D was also positively correlated with QUICKI (r = 0.24, P < .01), but was inversely correlated with HbA(1c) (r = -0.23, P < .01). Hypovitaminosis D was identified in 74% of obese subjects, whereas vitamin D deficiency was observed in 32.3% of our cohort. Vitamin D status was influenced by vitamin D intake, season, ethnicity/race, and adiposity. Interrelationships between 25(OH) D, iPTH, and FM were not influenced by season and race/ethnicity. Furthermore, serum 25(OH) D was positively correlated with insulin sensitivity, which was FM mediated, but negatively con-elated with HbA(1c), implying that obese children and adolescents with low vitamin D status may be at increased risk of developing impaired glucose metabolism independent of body adiposity. Additional studies are needed to evaluate the underlying mechanisms. (C) 2008 Elsevier Inc. All rights reserved.