LAUSR

Gonadotropin releasing hormone agonists (GnRHas) have been found to impair memory in adults, so the study by Wojniusz et al. (2016) on the possible cognitive effects of these drugs on children treated for idiopathic central precocious puberty (CPP) represents an important contribution to research in this area. Recent findings that GnRHas increase depression symptoms (Macoveanu et al., 2016) and slow reaction time (Stenbaek et al., 2016) in healthy women, and reduce long-term spatial memory in sheep (Hough et al., 2017) underline the importance of the research that Wojniusz et al. (2016) have undertaken. However, their reassuring statement in the abstract that girls undergoing GnRHa treatment for CPP and controls " showed very similar scores with regard to cognitive performance " and their conclusion that " GnRHa treated girls do not differ in their cognitive functioning ... from the same age peers " (Wojniusz et al., 2016) may be overly optimistic. These statements minimize the fairly substantial difference found in IQ scores and may also overemphasize its lack of statistical significance, as given the small number of participants in the study statistical significance has a high threshold. The statements should be qualified to indicate that the research has, in fact, reinforced concerns over the impact of GnRHas on cognitive performance in children. Girls treated for CPP with triptorelin acetate were tested with the short form Wechsler Intelligence Scale for Children III. It was found that the girls had a mean IQ of 94, as against a mean IQ of 102 for the matched control group (Wojniusz et al., 2016). These IQ estimations are presented as standardized IQ scores, which places a girl scoring 102 at the 55th percentile, and a girl scoring of 94 at the 34th percentile. It is questionable whether scores that indicate a percentile gap of this size can be described as " very similar. " The 8 point gap is not statistically significant (p = 0.09) but, as the authors point out, this may be a function of the small number of participants (15 treated girls, 15 controls). The authors contend that despite the small number of participants the results can—probably— be relied on to indicate that if GnRHas do cause a decline in IQ, this decline will be under 1 standard deviation (SD), which " represents a boundary of what is a clinically interesting difference "