Answers are grounded in this paper and cite their sources.
We thank Kushnir and coworkers1 for addressing the subject of how estradiol (E2) is used clinically in pediatric endocrinology. As Kushnir et al2 mention in the introduction of their original article, high-sensitivity methods for accurate measurements of E2 concentrations are necessary for the diagnosis of sex hormone–related disorders, precocious puberty (PP), and E2 deficiency and for determining the strategy for antiestrogenic treatment. Although PP is a clinical diagnosis, the laboratory testing for sex hormones in children suspected of having PP is performed not only for confirming the diagnosis but also, most importantly, to help diagnose the origin of the condition or disease and for the treatment strategy. For girls with PP with prepubertal E2 levels Figure 1, the origin can be exogenous estrogens or an ovarian cyst that has produced transient estrogens. PP with low E2 levels in the early puberty range (Figure 1) (in combination with increased luteinizing hormone [LH]/follicle-stimulating hormone [FSH]) can be an indication of a low tempo of the pubertal development and may not need treatment with a long-acting gonadotropin releasing hormone agonist, but patients will need close follow-up. PP with high E2 levels in the midpubertal range with increased LH/FSH needs treatment and further investigation of the origin of PP, eg, magnetic resonance imaging of the central nervous system for tumor. These are examples from routine clinical practice showing how important it is to have a reliable laboratory method for determination of E2 that can not only separate prepubertal E2 levels from pubertal levels but also elucidate whether the E2 levels are low or high in the pubertal stage. The liquid chromatography–tandem mass spectrometry (LC MS/MS) method described by Kushnir and coworkers2 is not useful clinically, with its wide overlap in any pubertal stages from prepubertal to pubertal and within pubertal Tanner stages. We agree with Kushnir and coworkers1 that all girls with elevated E2 levels do not have PP. Puberty starts in girls with a nocturnal increase of LH/FSH3,4 followed by a morning increase of E2,4 and it is the increased E2 levels that induce breast development (Tanner stage 2), which is usually the first physical sign of onset of puberty. Thereby, girls can have increased morning levels of E2 without signs of breast development. However, the levels are in the lower range of the Tanner stage 2 in the early pubertal range5 and not at all as high as in the reference for prepubertal girls shown by Kushnir and coworkers.2 This observation is further support that the LC MC/MC assay described by Kushnir and coworkers is not useful clinically.2 The 24-hour secretion pattern of serum estradiol in 74 healthy prepubertal and pubertal girls. Reference values at Tillväxtlaboratoriet (Göteborg, Sweden) depicted as the 95% confidence interval for the median in each pubertal stage. Kushnir and coworkers are well aware that there is a diurnal variation of sex steroids, but in their study they found a statistical difference between morning and afternoon serum E2 concentrations only in samples collected from boys in Tanner stage 3. Whether this is due to the method or to the study setting that the difference was not seen, we can only speculate. Unfortunately, owing to a typographic error, the data for boys are missing and the data for girls are instead shown twice (Figure 1B in the authors’ reply1 [correction published in Am J Clin Pathol. 2009;132:148]). Nevertheless, it is well established that girls during puberty have a diurnal pattern of E2 secretion (Figure 1),4,6,7 and so do boys.8,9 Regarding the concern of Kushnir and coworkers1 about the effect of ethnicity on our normal reference values,6,9 they were determined in whites and can be used as reference values for the white population in the world. Whether they are also useful for black/African American and Asian children needs to be studied. As long as there are no studies showing a difference in E2 secretion for white, black/African American, or Asian children, our reference intervals can be used for a wider population until we have ethnicityspecific references. Furthermore, our mean levels of E2 in the different pubertal stages for girls and boys are similar to what Klein et al10 found in the US population with their bioassay, lending further support that our reference values for whites are also useful for diagnostic use in diverse population groups, allowing for genetic and environmental differences. Kushnir and coworkers1 have concern regarding our reference values because the children were hospitalized for 2 days. Yes, the children were hospitalized for 2 days, but they stayed only 1 night in the hospital. The children arrived in the morning, and, therefore, we have 2 morning E2 values, 1 following the night when the child slept at home and 1 when the child slept at the hospital. There was no difference between these 2 determinations of morning E2 levels. Our reference values are representative for the type of samples obtained in the clinical setting to diagnose pubertal disorders. As mentioned in our previous letter to the editor,11 our reference values have been used in routine clinical practice for more than a decade, with clinically useful and reliable results. The samples admitted to our laboratory originate from outpatients of various ethnicities living all over Sweden. The imprecision data for the LC-MS/MS were reevaluated after the original article, this time with samples containing E2 concentrations in the low range.1 However, it is not clear what the samples originated from. The functional sensitivity is not evaluated properly until pure child serum samples (not diluted or spiked) of low E2 concentrations have been determined. It is a challenge to determine E2 values in low concentrations, and it is established that direct immunoassays do not perform well, not even those with low analytic detection limits.12–15 The intention with the work by Kushnir et al2 was to develop a high-sensitivity LC-MS/MS assay for serum E2 levels. It is therefore surprising that Kushnir et al2 refer to the reference values by Wennink et al3 for prepubertal E2 values in girls. The E2 direct radioimmunoassay by Wennink et al3 had a 10 times higher analytic detection limit (10 pg/mL = 37 pmol/L) compared with that of Kushnir and coworkers2 (1 pg/mL = 4 pmol/L), the cross-reactivity for estrone and estriol was 6%, and the interassay variation ranged between 13% and 33%. Despite the 10-times difference in the lower detection limits between these methods, the median values for prepubertal girls are comparable, according to the authors’ reply.1 In our opinion, at present, the tandem mass spectrometry assay method described by Kushnir and coworkers2 is not useful clinically. However, we believe that the future of steroid determination will be in the mass spectrometry assay, but we are not there yet.
No takes yet. Share an insight, caveat, or question.
Ankarberg‐Lindgren et al. (2009) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: