| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Letters to the Editor |
University of Florida Gainesville, Florida 32611-7047
We welcome the opportunity to reply to the wide-range of comments about our paper from the pioneering investigators of this condition. To begin with, we do not understand why Laron and Galatzer (in the letter above) refer to GH receptor deficiency (GHRD)/Laron syndrome as primary IGF-I deficiency. Such terminology is correctly applied to intrinsic failure of an endocrine organ (e.g. thyroid agenesis with primary hypothyroidism) or to genetic defects in hormone synthesis, such as the defect in the IGF-I gene recently described by Woods et al. (1). In contrast, IGF-I deficiency secondary to GHRD is, obviously, not primary. Equating GHRD with primary IGF-I deficiency is equivalent to referring to TSH resistance as primary thyroid hormone deficiency (2). The classification reference provided by Laron and Galatzer (3) does not, in fact, use the term primary IGF-I deficiency, although it illogically includes primary defects in IGF-I synthesis, such as that described by Woods et al. (1), under the rubric Primary GH insensitivity syndromes (Laron syndrome).
The Israeli group did, indeed, report low IQ in most of their patients with Laron syndrome but, in contrast to our studies, they did not control their observations by testing unaffected siblings or unrelated control subjects from the same ethnic community. IQ tests have long been criticized for being culture- and language-specific (4). Given the cultural heterogeneity of the Israeli population, it would be particularly important to have appropriate controls for a patient group that originates exclusively from Middle Eastern and North African communities. Underestimation of true intelligence in some of their subjects is suggested by the remarkable achievement of a Ph.D. in microbiology by one of their patients who had a measured IQ of 106. An IQ of 73 in one sibling and 46 in another is hardly an example of "great variability." Because the standard deviation of IQ for offspring of the same parents raised together is approximately 14 points (5), a difference of 27 points between siblings is well within the limits of normal intrafamilial variability.
The promise of "a complete correlation between psychological abilities and molecular defects" in Laron syndrome is naive, as well as inconsistent with the belief of Laron and Galatzer that there is an important influence of GH-dependent IGF-I synthesis on brain function. If that were so, it would not matter what mutation caused the GHR failure.
Laron and Galatzer seriously misread Table 2 in stating that our conclusions do not fit the data presented. We administered a battery of four chronometric and two psychometric tests of intelligence that were designed to minimize the effects of physical size, motor coordination, and cultural background on test performance. While noting that the community controls had a significantly higher mean score on the psychometric tests than did the patients with GHRD, Laron and Galatzer overlooked the facts that (a) patients with GHRD and community controls performed comparably on the chronometric tests; and (b) community controls also had significantly higher mean scores on the psychometric tests than did unaffected relatives of the patients. Most importantly, they have ignored the absence of significant differences between patients with GHRD and their unaffected relatives across the entire battery of chronometric and psychometric tests used in our study, with the single exception of one task (viz. Inspection Time), on which the subjects with GHRD performed better than did their unaffected relatives. This pattern of results is consistent with the sampling bias we suspected as a result of our inability to select control subjects randomly from the schools that the children attended; we had to rely on referral by teachers, who were likely to select their best students.
Laron and Galatzer are also incorrect in stating that the scoring of our intelligence tests differed by age. Norms are not needed for interpretation of the chronometric tests because performance is measured in milliseconds by computer (i.e. on a ratio scale of measurement). Norms were also not used to interpret the psychometric test results, because norms for these tests were not developed in Ecuador. Thus, we used the simple unit-weighted sum of the raw scores on the two tests (after transformation to z-scores) to provide one estimate of psychometric intelligence, possible because of the high correlation observed between tests. Age differences between groups were statistically controlled in all analyses. The scoring of our chronometric and psychometric tests of intelligence, therefore, did not differ by age.
Laron and Galatzer also suggest that the variability in IQ scores of subjects with GHRD in Table 2 casts doubt on our conclusions. Although we did not report on differences in variability across groups in our paper, examination of F ratios indicates that the variability of performance on the chronometric and psychometric tests of IQ for the patients with GHRD is not significantly different than that for their relatives or for the community controls.
There was no reason to include information on head size or any other previously reported auxological data in our paper. It is not true, however, that there is no information, "regarding head circumference in the Ecuadorian cohort, nor ... on the early motor skills in infancy," in the review by Rosenfeld et al. (6), which states that there is "normal head size and forehead length for stature," based on analyses of the Ecuadorian patients and those from the literature (7). The review further notes that delayed walking was reported in 70% of Ecuadorian patients, presumably because of hypomuscularity, and that this was similar to observations made initially by the Israeli group. In contrast to the Israeli experience, however, Ecuadorian subjects had normal timing of speech onset (6, 7, 8, 9).
Laron and Galatzer note that "small head (brain) increases rapidly with IGF-I replacement treatment." The reference provided, their 1988 study of the effect of single intravenous bolus injections of IGF-I in nine patients, does not contain any such data, and it would be truly amazing if it did! Their later report of a remarkable effect of 10 months of IGF-I replacement therapy on head growth in two patients (10) was not followed up in their longer-term treatment report, involving nine patients treated for 1 yr and six of them for 2 yr. This later report includes no data on head growth, nor have the Israeli investigators (or anyone else) suggested improvement in intellectual performance with IGF-I replacement therapy (11).
Laron and colleagues have presented no controlled data that point to an important role for GH-dependent IGF-I in intrauterine and postnatal brain growth and function. Their cohort of patients with Laron syndrome includes individuals with very low intelligence and those comparably deficient in IGF-I who are above normal in intelligence or achievement. In a population such as this, with a high rate of consanguinity, an increased frequency of cognitive impairment would not be unexpected. Sporadic reports of individuals or groups of patients with GHRD suggest that the distribution of intelligence is normal (8, 9), and our own controlled studies indicate no difference between affected individuals and their unaffected siblings. The explanation of these observations is that function of the GHR is not necessary for normal cognitive development. This does not necessarily mean that IGF-I is unnecessary for brain development; the significantly higher circulating concentrations of IGF-I and IGFBP-3 in adults with GHRD compared with affected children suggests alternative mechanisms for generating IGF-I than via the GH-GHR (6, 12).
Nothing that the Israeli group has published or that is stated in their letter argues against our conclusion that "GH-induced IGF-I production is not required for normal brain growth in utero or for postnatal intellectual development (13)."
Footnotes
Address correspondence to: John H. Kranzler, Department of Educational Foundations, 1403 Norman Hall, P.O. Box 117047, University of Florida, Gainesville, Florida 32611-7047.
Received August 24, 1998.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
