Correspondence
Asthma, Corticosteroids, and Growth
N Engl J Med 2001; 344:607-608February 22, 2001
- Article
To the Editor:
The studies by the Childhood Asthma Management Program Research Group1 and by Agertoft and Pedersen2 (Oct. 12 issue) suggest that the magnitude of the growth effect observed during the first year of therapy with inhaled corticosteroids in children with asthma tends not to be sustained during subsequent years. However, neither study was adequately designed or had sufficient power to assess the impact of inhaled corticosteroids on final adult height. Moreover, reporting height data as means ±SD precludes the identification of more seriously affected children with outlier values.
Agertoft and Pedersen2 describe an open-label, nonrandomized study without concurrent control that provided final heights for less than 50 percent of the original 332 patients. An estimate of adult height based on mean parental height was introduced post hoc as the primary comparison. Given the substantial effect of the method of height prediction on estimates of final adult height, in addition to wide confidence intervals of ±10 cm,3 the authors' conclusions appear to be overstated.
The primary objective of the Childhood Asthma Management Program study1 was to assess lung growth in children with asthma, and the second objective was to assess growth velocity. Because of concern that inadequately controlled childhood asthma may ultimately compromise lung function in adulthood,4 the finding of no significant difference in lung function is noteworthy. The children treated with inhaled corticosteroids had a mean decrement in growth velocity of 1 cm during the first year. Subsequently, there was a gradual return toward the rates observed in the cohort not treated with inhaled corticosteroids, although with a notable absence of “catch-up” growth or evidence of a lengthening of the prepubertal growth period. Therefore, the study fails to explain how height lost early in treatment might be regained. Although the predicted adult height may have been reached, it is not synonymous with the potential adult height.
(The views expressed are those of the authors and do not necessarily represent those of the Food and Drug Administration or imply its endorsement.)
4 ReferencesMary Purucker, M.D., Ph.D.
Saul Malozowski, M.D., Ph.D.
Food and Drug Administration, Rockville, MD 208571
The Childhood Asthma Management Program Research Group
Full Text | Web of Science | Medline2
Agertoft L ,Pedersen S . Effect of long-term treatment with inhaled budesonide on adult height in children with asthma. N Engl J Med 2000;343:1064-1069
Full Text | Web of Science | Medline3
Luo ZC ,Low LCK ,Karlberg J . A comparison of target height estimated and final height attained between Swedish and Hong Kong Chinese children. Acta Paediatr 1999;88:248-252
CrossRef | Web of Science | Medline4
Agertoft L ,Pedersen S . Effects of long-term treatment with an inhaled corticosteroid on growth and pulmonary function in asthmatic children. Respir Med 1994;88:373-381
CrossRef | Web of Science | Medline
To the Editor:
I have serious reservations about the conclusions reached by the Childhood Asthma Management Program Research Group. The study design allowed beclomethasone or “other asthma medications” to be given to children in all three groups. Not surprisingly, the children who received budesonide used significantly less beclomethasone or other medications than those in the placebo group. Thus, the study did not really compare the budesonide group with a placebo group but, rather, compared the budesonide group with a group in which a substantial percentage of the children received another aerosol steroid. In previous studies, beclomethasone has been shown to lower growth velocity in children with asthma.1
The minor reduction in growth velocity seen in the budesonide group as compared with the placebo group might have been much larger. Conversely, the lack of effect on forced expiratory volume in one second in the budesonide group might also have been an artifact of an inadequate “placebo” group. I understand that the use of a true placebo in such a long longitudinal study might have been deemed unethical and the need for an alternate “rescue” aerosol steroid a necessity. Nevertheless, one has to analyze the results of the study carefully, given this flaw in the design.
1 ReferencesJonathan Ilowite, M.D.
Winthrop University Hospital, Mineola, NY 115011
Simons FER, Canadian Beclomethasone Dipropionate-Salmeterol Xinafoate Study Group
Full Text | Web of Science | Medline
Author/Editor Response
The authors reply:
To the Editor: Although height was not the primary outcome in the randomized trial conducted by the Childhood Asthma Management Program, it was a prespecified secondary outcome. The sample size (1041 children) and length of follow-up (mean, 4.3 years) far exceed those of earlier trials demonstrating a growth effect of inhaled or nasal beclomethasone dipropionate.1-4
The effect on height after 4.3 years of treatment was a reduction in growth of 1.1 cm (P=0.005), occurring mostly within the first year, in the budesonide group as compared with the placebo group. This effect is much smaller than the reduction of 1 cm per year suggested by other studies.1-4 The absence of a longer-term effect on height may be due to compensatory mechanisms that limit the effect of the steroid, as has been shown with bone density in patients receiving higher doses of inhaled steroids.5 Adult height could not be measured, but predicted adult height, derived with use of standard methods to extrapolate from current height, current age, and bone age, showed no difference among the groups (it is not clear how “potential adult height” could be determined).
The regression analysis suggested by Dr. Ilowite showed that the use of beclomethasone during the trial did not significantly modify the effect on height or explain the lack of an effect on lung function (the results of this post hoc analysis are subject to bias since they are not from the intention-to-treat analyses). Furthermore, the significantly lower airway responsiveness in the budesonide group argues against the suggestion that the observed effects were simply due to the use of beclomethasone in the placebo group. Although means and standard deviations may not permit the identification of more seriously affected outliers, supplementary analyses demonstrated that extreme changes in height were no more likely to occur in the budesonide group than in the placebo group. It remains important that physicians treating children with inhaled corticosteroids monitor growth regularly and adjust doses downward if growth is affected. Other measures of potential effects of steroids (bone density, Tanner stage, bone age, results of eye examinations, etc.) did not provide support for any significant systemic adverse effect. There is obviously individual variation in beneficial and adverse responses, which our results do not address.
5 ReferencesStanley J. Szefler, M.D.
Scott Weiss, M.D.
James Tonascia, Ph.D.
Johns Hopkins University, Baltimore, MD 21205for the Childhood Asthma Management Program Research Group
1
Tinkelman DG ,Reed CD ,Nelson HS ,Offord KP . Aerosol beclomethasone dipropionate compared with theophylline as primary treatment of chronic, mild to moderately severe asthma in children. Pediatrics 1993;92:64-77
Web of Science | Medline2
Doull IJM ,Freezer NJ ,Holgate ST . Growth of prepubertal children with mild asthma treated with inhaled beclomethasone dipropionate. Am J Respir Crit Care Med 1995;151:1715-1719
Web of Science | Medline3
Simons FER, Canadian Beclomethasone Dipropionate-Salmeterol Xinafoate Study Group
Full Text | Web of Science | Medline4
Skoner DP ,Rachelefsky GS ,Meltzer EO , et al. Detection of growth suppression in children during treatment with intranasal beclomethasone dipropionate. Pediatrics 2000;105:E23-E23
CrossRef | Web of Science | Medline5
Toogood JH ,Baskerville JC ,Markov AE , et al. Bone mineral density and the risk of fracture in patients receiving long-term inhaled steroid therapy for asthma. J Allergy Clin Immunol 1995;96:157-166
CrossRef | Web of Science | Medline
To the Editor:
Purucker and Malozowski raise four issues that relate to our study. We agree that the presentation of height data as means ±SD precluded the identification of outliers. An unpublished figure that included the individual height percentiles of all patients before and after treatment with budesonide showed that if a child was tall for his or her age before treatment, he or she would also end up tall as an adult, and vice versa. There were also no more outliers after than before treatment.
We do not know what the claim about inappropriate study design and insufficient power is based on. Our study had a high power to detect differences between measured and target adult heights and between adult heights of children treated with budesonide and those of their siblings (or siblings and controls). Although a randomized, double-blind design might have been better, it was not possible in a 15-year study in which big differences in the dropout rates between the groups in a study population similar to ours would have complicated the analysis. Furthermore, an additional analysis showed that there was no reason to suspect that the results for the children who had not yet reached adult height would prove to be any different from the results for the children who had reached adult height.
It is not correct that the primary comparison was a post hoc comparison. For adult height, the protocol stated that adult height based on mean parental height was the primary outcome.
We based our conclusion not only on the finding that the children grew to be taller than their parents and 95 percent reached an adult height within 8 cm of their target height, but also on the finding that the children treated with budesonide achieved their target height to the same extent as their healthy siblings and the 18 controls with asthma who had not received any inhaled steroids. The strong correlation between adult height and the standard-deviation score for height before treatment with budesonide provided additional support. The wide confidence interval for the prediction of adult height in the literature did not weaken our conclusion.
Lone Agertoft, M.D.
Søren Pedersen, M.D., Dr.Med.Sci.
Kolding Hospital, DK-6000 Kolding, Denmark- Citing Articles (3)
Citing Articles
1
J. Zhou, D.-F. Liu, C. Liu, Z.-M. Kang, X.-H. Shen, Y.-Z. Chen, T. Xu, C.-L. Jiang. (2008) Glucocorticoids inhibit degranulation of mast cells in allergic asthma via nongenomic mechanism. Allergy 63:9, 1177-1185
CrossRef2
Giovanni A. Rossi, Franklin Cerasoli, Mario Cazzola. (2007) Safety of inhaled corticosteroids: Room for improvement. Pulmonary Pharmacology & Therapeutics 20:1, 23-35
CrossRef3
(2001) Current Awareness. Pharmacoepidemiology and Drug Safety 10:4, 345-360
CrossRef
