Background: Subcutaneous gonadotropin-releasing hormone (GnRH) analog test is an alternative to standard intravenous GnRH test for diagnosing central precocious puberty (CPP). Previous studies showed different cutoffs of basal and peak serum LH levels following intravenous GnRH and subcutaneous GnRH analog for diagnosing CPP girls. However, no study that defined the discriminating basal and peak LH levels among CPP and premature thelarche (PT) girls with different breast Tanner stages, is available.
Objective: To determine basal and GnRH analog-stimulated peak LH levels for differentiating girls with PT and CPP at breast stages II and III.
Methods: Medical records of 515 girls with breast onset before 8 years of age who underwent subcutaneous GnRH analog test between the years 2007 and 2017 were reviewed. Girls who had progressive breast development and accelerated growth during a 36 month period of follow-up, advanced bone age and pubertal-sized uterus and ovaries were diagnosed with CPP. The girls who had no above-mentioned findings were classified as having PT. In each group, patients were divided into 2 groups according to their breast stages at the time of GnRH analog testing, Tanner II (CPPII and PTII) and Tanner III (CPPIII and PTIII). The GnRH analog test results were analyzed.
Results: Of 515 girls, there were 121, 126, 211 and 57 girls with CPPII, PTII, CPPIII and PTIII, respectively. Their median (IQR) ages at the breast onset were 7.6 (7.2, 7.9), 7.4 (6.8, 7.8), 7.6 (7.0, 7.9) and 7.1 (6.3, 7.6), respectively. Basal and peak LH levels of 0.11 and 5.02 IU/L, respectively provided the sensitivity and specificity of 72% and 82%, and 79% and 91%, respectively in diagnosing CPP. Among the 4 groups, basal and peak LH levels were highest in CPPIII girls [median (IQR) basal LH: 0.24 (0.11, 0.63) IU/L, P<0.001; peak LH: 9.53 (5.92, 8.35) IU/L, P<0.001]. Subgroup analysis revealed that basal LH cutoff of 0.21 IU/L had 38% sensitivity and 94% specificity in discriminating CPPII from PTII while it provided a sensitivity of 56% and a specificity of 97% in differentiating CPPIII from PTIII.
Conclusion: A single basal serum LH level of greater than 0.21 IU/L can be used for diagnosing girls with CPP. With this basal LH level, CPP can be diagnosed without GnRH analog test in approximately 38% and 56% of girls with Tanner II and III, respectively.
27 - 29 Sep 2018
European Society for Paediatric Endocrinology