Cross-Sectional and Longitudinal Associations between Body Mass Index and Cardiometabolic Risk Factors in Adolescents in a Country of the African Region

Abstract

We assessed the association between several cardiometabolic risk factors (CRFs) (blood pressure, LDL-cholesterol, HDL-cholesterol, triglycerides, uric acid, and glucose) in 390 young adults aged 19-20 years in Seychelles (Indian Ocean, Africa) and body mass index (BMI) measured either at the same time (cross-sectional analysis) or at the age of 12–15 years (longitudinal analysis). BMI tracked markedly between age of 12–15 and age of 19-20. BMI was strongly associated with all considered CRFs in both cross-sectional and longitudinal analyses, with some exceptions. Comparing overweight participants with those having a BMI below the age-specific median, the odds ratios for high blood pressure were 5.4/4.7 (male/female) cross-sectionally and 2.5/3.9 longitudinally (P < 0.05). Significant associations were also found for most other CRFs, with some exceptions. In linear regression analysis including both BMI at age of 12–15 and BMI at age of 19-20, only BMI at age of 19-20 remained significantly associated with most CRFs. We conclude that CRFs are predicted strongly by either current or past BMI levels in adolescents and young adults in this population. The observation that only current BMI remained associated with CRFs when including past and current levels together suggests that weight control at a later age may be effective in reducing CRFs in overweight children irrespective of past weight status.

Childhood obesity and adult cardiovascular disease risk factors: a systematic review with meta-analysis

“…The year that each study started varied considerably, ranging from 1923 to 1989 while the year that studies were published ranged from 1971 to 2014. Studies were conducted in eleven different countries; six in the U.S. [ 4 , 64 , 66 , 67 , 73 , 75 ], three in United Kingdom (England, Wales, Scotland and Newcastle) [ 76 , 83 , 85 ], three in Finland [ 65 , 68 , 81 ], two in Australia [ 77 , 82 ], and one in either Sweden [ 69 ], India [ 71 ], Lithuania [ 72 ], Poland [ 74 ], the Republic of Seychelles [ 78 ], Japan [ 79 ], or the Solomon Islands [ 84 ]. Most studies used a prospective longitudinal study design except for two studies that used a retrospective study design [ 65 , 73 ].…”

“…Sixteen studies examined the association between childhood obesity and adult SBP [ 4 , 64 , 66 – 69 , 71 – 73 , 75 – 79 , 82 , 85 ], 14 examined the association between childhood obesity and adult DBP [ 4 , 64 , 67 – 69 , 71 , 72 , 75 – 79 , 82 , 85 ], 8 examined the association between childhood obesity and adult TC [ 4 , 64 , 67 , 79 , 81 , 83 – 85 ], 5 examined the association between childhood obesity and adult LDL [ 4 , 78 , 81 , 83 , 85 ], 8 examined the association between childhood obesity and adult HDL [ 4 , 67 , 68 , 78 , 81 – 83 , 85 ], and 9 examined the association between childhood obesity and adult TG [ 4 , 67 , 68 , 71 , 78 , 81 , 82 , 84 , 85 ]. However, data from one study that assessed TG could not be used for meta-analysis because the study’s main outcome was blood pressure and there was not enough information provided for TGs to calculate an ES [ 71 ].…”

“…Another study by Holland and colleagues was excluded from the meta-analysis for the same reason [ 70 ]. Of the 21 studies, only six (28.6%) included data on the association between childhood obesity and adult CVD risk factors while adjusting for adult BMI [ 4 , 65 , 74 , 76 , 78 , 85 ]. Two of these six studies reported adjusted associations only [ 65 , 74 ].…”

“…No study was limited to women. One study included combined results as well as results according to sex [ 78 ]. However, for the current meta-analysis, we used the results reported according to sex [ 78 ].…”

Inverse association between height increase and LDL cholesterol during puberty: A 3‐year follow‐up study of the Fukuroi City

“…It is well known that serum LDL‐C is positively associated with body weight among children (Azita et al, ; Lyngdoh et al, ; Shirasawa et al, ). Serum LDL‐C is influenced by both height and weight.…”