Abstract

OBJECTIVE:

To compare the performance of recently released body mass index percentiles (BMIp) with standard anthropometric indexes, including height-for-age percentile (HAP), weight-for-age percentile (WAP) and percent ideal body weight (%IBW), as measures for nutritional failure in children with cystic fibrosis (CF).

DESIGN:

Cross-sectional analysis of growth and lung function data from 4577 children with CF reported to the German CF quality assurance (CFQA) project from 1995 to 2004.

RESULTS:

Frequency distribution of HAP (mean+/-s.d.: male 30.0+/-27.5; female 31.3+/-27.4) and WAP (male 28.9+/-27.0; female 29.6+/-26.7) were skewed, with significant numbers of patients below the fifth percentiles of a healthy reference population. However, because deficits occurred in both measures simultaneously, mean %IBW (male 97.0+/-12.1; female 98.1+/-12.3) assumed subjects weight close to the nominal weight-for-height at all ages. In contrast, mean BMIp was markedly reduced (male 35.7+/-27.9; female 35.6+/-27.2) and steadily declined with age. Ideal weight-for-age was significantly lower when predicted by %IBW compared with BMIp method, particularly in subjects with shorter-than-average stature. Consequently, less CF children were identified with nutritional failure according to %IBW method (male 20.5%; female 22.7%) compared with BMIp method (male 30.4%; female 28.7%). The clinical relevance of these findings was confirmed by stronger correlation of BMIp with impaired %forced expiratory volume/s, a marker for disease progression in CF.

CONCLUSION:

BMIp predicts nutritional failure more sensitively and accurately than conventional anthropometric indexes, at least in children with CF. Screening of CF patients by BMIp could provide an early warning sign and allow for timely therapeutic intervention.