Calculation of knemometric growth rates in group studies of children treated with exogenous glucocorticoids

Calculation of growth rates in short-term knemometry studies is often based on linear regression analysis. The method, however, has not been compared to the conventional method for calculation of growth rates (increase in length divided by the time interval) in glucocorticoid-treated populations. The aim of the present study was to assess whether the conventional method for calculation of growth rates is as sensitive as the linear regression method in the assessment of short-term lower leg growth in group studies of children treated with exogenous glucocorticoids. Three different studies were conducted during which knemometry was performed once or twice weekly. Six boys and six girls aged 5-12 years, with atopic dermatitis, participated in an open longitudinal study with three periods of 2 weeks duration. Ten boys and four girls aged 12-16 years, with asthma, were studied in a randomized double-blind, placebo-controlled two-period crossover trial, with treatment periods of 4 weeks. Thirteen boys and 11 girls aged 5-12 years, with asthma, were enrolled in a randomized double-blind two-period crossover trial with treatment periods of 4 weeks. During active treatment periods topical budesonide was given. Mean lower leg growth rates and the results of significance tests based on the two methods for calculation of growth rates were similar in all studies. Significant correlations between the growth rates based on the two methods were found in the non-glucocorticoid and glucocorticoid treatment periods in all studies (Pearson correlation r coefficient varied from 0.81 to 0.97, p from <0.001 to < 0.0001). The conventional method for calculation of lower leg growth rates is as reliable as linear regression analysis. By using the conventional method the number of visits required for knemometric growth rate assessment in group studies of systemic activity of exogenous glucocorticoids can be reduced.