[Are they valid Spanish growth reference charts?]

To modify the results of the longitudinal study on growth in Navarra (NA 09) by censoring the obesity cases from the population sample as well as to perform a comparative analysis with the most qualified Spanish and international growth curves.

all the cases with obesity according to Cole et al. standards have been censored out of the 930 participants in NA-09, the final sample being 782 participants (371 males and 411 females). The results obtained were compared with the Spanish studies by Serra-Majem et al. (enKid study), Carrascosa et al. (ESP 08), and NA 09, which do not censor the obesity cases, and with the Centers for Disease Control and Prevention (CDC, 2000) table, and the WHO tables (WHO, 2007) that apply depuration criteria of the poorly healthy anthropometric data (obesity and malnutrition).

We present the mean values adjusted by height, weight, and BMI with their percentile distribution for both genders. When comparing with the Spanish studies, we observe that the evolutionary values of the 3d and 50th percentiles for height, weight, and BMI are virtually the same; however, the evolutionary values for the 97th percentile for weight and BMI tend to differ more and more. When comparing them to the international standards, the evolutionary values for the 3d, 50th, and 97th percentiles for BMI lay between both references.

for the growth curves and tables to be useful as reference patterns, all obese people should be excluded from their elaboration; otherwise, they should be considered as only descriptive studies of a population with a recognized tendency to excessive body weight and thus their clinical applicability would be put in question.