Reliable indices that are consistently able to predict the linear growth promoting effects of recombinant human growth hormone (rhGH) in short children have long been sought.  The authors analyzed data from a National Cooperative Growth Study of the usefulness of IGF determinations, auxological measurements, and 12-hour serial GH measurements obtained every 20 minutes between 2000 and 0800 hours, in children who were treated with rhGH (0.29 mg/kg/week in 6+ weekly injections) for a mean of 3.6-3.8 years. There were 825 prepubertal children with short stature studied (mean height -2.8 SDS; bone age delay ~2.3 years).  The children were subdivided into one group of 300 (231 males, 69 females) with isolated GH deficiency (IGHD - peak GH response to provocative stimulation <10 ng/mL by unstated methods) and 525 (404 males, 121 females) with idiopathic short stature (ISS - peak stimulated GH response >10 ng/mL).  The data were analyzed by the cluster program.  In addition, a measurement of the “orderliness” or “regularity” of overnight spontaneous, endogenous GH secretion, which is termed “approximate entropy”, was calculated.

As anticipated, mean and maximum spontaneous peak GH levels, pooled mean GH concentrations, and mean area under the GH peaks were significantly lower in subjects with IGHD than in those with ISS.  Interestingly, pretreatment IGF-I concentrations were similar in the two groups (120 and 125 ng/mL, respectively).  The increment in height SDS after treatment with rhGH was similar in the two groups (+1.2 to 1.3 SDS).  Significant but weak correlations (r <0.4) related rhGH-induced height increment to height deficit prior to treatment, duration of treatment, and mid parental height SDS in both groups.  Maximum stimulated GH values, spontaneous overnight GH measurements, and pre-treatment IGF-I levels were also inversely related to rhGH-induced growth, but again the r values were low (-0.15 to -0.395).  By multiple regression analyses, only the peak GH response to secretagogue was inversely correlated to treatment related height increment; spontaneous GH measurements were not related.  When data from children with “severe” IGHD (peak stimulated GH response <5 ng/mL) or “extreme” ISS (height <-3.3 SDS) were isolated and examined, spontaneous GH measurements were inversely related to treatment induced growth but did not improve calculated height prediction models.  Spontaneous GH secretion was more orderly in children with severe IGHD than those with “moderate” IGHD.  In the ISS subjects, GH secretion was more orderly in those with “mild” ISS, and IGF-I concentrations were higher than in those with extreme ISS.  The authors conclude that in general, serial measurements of spontaneous overnight GH secretion did not provide information helpful in the prediction of the linear growth response to rhGH, thus supporting the conclusion of several earlier studies of this question.

Rogol AD, et al.  Clin Endocrinol 2003;58:229-237.

First Editor’s Comment: Clearly, the diagnosis of IGHD is fraught with difficulty as 40-80 percent of such children will have normal GH secretion as adults.  Thus, there must be overlap between the diagnostic categories of IGHD and ISS in this study.  In this regard it is of interest that the “disorderliness” of GH secretion was greatest in those subjects with “moderate” IGHD and both “mild” and “extreme” ISS – implying a close relationship between these groups in the regulation of GH secretion.  As the investigators suggest, it may be that a defect in the “orderly” secretion of GH is translated into decreased tissue responsiveness to GH even though the absolute amount of GH secreted may be normal.  Although several factors were related to height increment on rhGH, none had the high r value we seek as an “absolute” predictor of response.  Hence, the search goes on!

Allen W. Root, MD

Second Editor’s Comment: I feel pressured to comment that a possible reason that many children appear to be GH deficient as children but not as adults is that the sex hormones stimulate GH release.  We use the same threshold criteria for GH release to secretagogues in adults as children.  How do we know that apparent isolated GHD children are not still partially GHD as adults?  Studies are needed in this group of patients when they reach adulthood to evaluate comparison of GH response to secretagogues in adults who were not thought to be GHD as children.  In such a study we might find that those diagnosed with GHD as children are still GHD as adults relative to others who were never short as children.  The fact that many pediatric endocrinologists used to prime suspect GHD children with testosterone before administering secretagogues for the purpose of exaggerating the GH response in suspect GHD children supports my hypothesis.

Robert M. Blizzard, MD

Third Editor’s Comment: In the previous issue of Growth, Genetics & Hormones (Vol. 19, No. 2) a paper by Lanes and Jacubowitz was reviewed.¹ These authors also showed that IGF-I and IGFBP3 were not useful in assessing the response to hGH therapy. Careful measurements and monitoring of growth are the gold standards.

Fima Lifshitz, MD

Reference

1. Lanes R, Jacubowitz S. J Pediatr 2002;141:606-610.