Knemometry is more sensitive to systemic effects of inhaled corticosteroids in children with asthma than 24-hour urine cortisol excretion

Abstract

Pharmacodynamic assessment of the systemic effect of inhaled corticosteroids (ICSs) is often done by measuring 24-hour urine free cortisol (UFC) excretion. Knemometry assessing short-term lower-leg growth rate (LLGR) is a more rarely used alternative. The primary aim of this study was to compare the sensitivity of LLGR and 24-hour UFC excretion for evaluating systemic exposure to ICSs in prepubertal children with asthma. The secondary aim was to evaluate factors influencing the precision of LLGR calculated by the traditional 1 leg nonparametric method versus a new 2 leg parametric method. The study evaluated 60 children with mild asthma aged 5 to 12years participating in a randomized controlled trial of ICSs with longitudinal concomitant assessments of LLGR and 24-hour UFC excretion. The sensitivity of the safety assessments was analyzed by comparing LLGR and 24-hour UFC in the placebo run-in period with values in the ICS treatment period by using paired t tests. Factors with a potential influence on LLGR were analyzed by means of ANOVA and the Levene test of homogeneity. The mean LLGR was significantly reduced during the ICS versus placebo run-in periods: 0.18mm/wk (SD, 0.55mm/wk) versus 0.45mm/wk (SD, 0.39mm/wk), with a mean difference of 0.27mm/wk (95% CI, 0.05-0.48mm/wk; P=.02). In contrast, there was no difference in 24-hour UFC excretion: 6.91nmol/mmol (SD, 4.67nmol/mmol) versus 7.58nmol/mmol (SD, 6.17nmol/mmol), with a mean difference of 0.67nmol/mmol (95% CI, -1.13 to 2.48nmol/mmol; P=.46). We observed no significant difference in parametric determined LLGR caused by the child's age or sex, investigator, or season of measurement, whereas some differences were observed for the nonparametric LLGR. These findings suggest that knemometry is a more sensitive pharmacodynamic measure of systemic effects of ICSs than 24-hour UFC excretion and that a parametric determination of LLGR increases the sensitivity of the method. These findings should be considered by legislative authorities in the future.