Growth charts: do they reflect healthy growth in Sri Lankan children?

Anthropometry has been used in the assessment of growth with "normal" range of growth being considered to span within ±2SD of median. Generally the practice is to make a child to grow within this standard deviation (SD) range. This study aims to investigate the hypothesis that all children with anthropometrical parameters within the designated "normal growth range" (within ±2SD of median), have a healthy body composition irrespective of the SD value.

Five to 15-year-old apparently healthy Sri Lankan children were studied. SD scores for height, weight and BMI, for age were calculated based on WHO 2007 standards. Sample was stratified according to age, sex, and SD score. %FM (assessed by isotope dilution methods using D2O) was compared between each SD group of each anthropometric parameter for gender and age. There were 278 (M/F:154/124) children. Non-parametric Kruskal-Wallis test revealed that the %FM of children differed significantly between SD groups even when they were within the accepted "normal limits of growth" and had higher levels of %FM in the upper SD groups.

This study shows that Sri Lankan children have unacceptably high body fat at levels that have been designated by growth charts as "normal". Children who are positioned at a higher SD score, +1 to +2SD, (within normal range) in the weight for age and BMI for age charts are mainly due to the deposition of fat rather than the growth in both fat and fat free compartments of body. Therefore the implication of using current cutoff values would be leading to the accumulation of excess fat in the body of those with higher SD values, but still considered to be within "normal" limits. Therefore in the absence of local standards, it would be important to revise the cutoff values of the WHO growth charts to suite Sri Lankan children which would be between -3SD and +1SD.