This is a retrospective monocentric study.
We recorded MRI data of 123 consecutive girls referred to the Pediatric Endocrinology Unit of University of Pavia, Italy, from January 2010 to December 2015 for early onset puberty. Both the history of the disorder and a physical examination of each girl suggested progressive pubertal development according to the Tanner classification. The patients were followed for at least three years after the end of therapy or completion of the follow up period.
Out of the patients, 100 had a CPP diagnosed according to the classic definition if the appearance of the breast development was before the age of 8 and 23 patients had a CEP defined if the appearance of the breast development was between eight and ten years .
For 71 patients it was possible to collect auxological, hormonal and ultrasonographic parameters.
One girl of Italian origin had been adopted; and 8 girls were non-Caucasian (4 South Americans, 3 Africans and 1 Asian).
Girls with dimorphic syndrome, chromosome abnormalities, endocrine or chronic diseases, neurological disorders or CNS pathologies were excluded from the study. None of the girls were overweight or obese, their BMIs being within normal range.
Height was measured using a Harpender stadiometer and expressed as standard deviation score (SDS). Body mass index (BMI) was calculated according to the formula weight (kg)/height2 (m2). Both height and BMI SDS were calculated according to the charts developed by Cacciari et al.  and were within normal range. Bone age was assessed using the Greulich and Pyle Atlas .
At time of diagnosis, all patients were evaluated through auxological data including pubertal stages, bone age, basal and peak serum levels of gonadotropins after the GnRH stimulation test, serum estradiol values, and both uterine and ovarian measurements by pelvic ultrasound. All subjects underwent the GnRH stimulation test (100 μg/m2 GnRH given i.v. bolus) to evaluate serum luteinizing hormone (LH) and follicle-stimulating hormone (FSH) determined by chemiluminescent immunometric assay (Siemens Medical Solutions Diagnostics, Milan, Italy). We considered stimulated LH levels > 5 IU/l as a diagnostic cut-off of pubertal range.
Thorough pelvic ultrasound using an Aloka Prosound SSD 5500 machine was performed to evaluate uterine length and transverse diameter, fundus/cervix ratio, ovarian volume, and the presence of an endometrial echo. Our study considered as pubertal size a uterine length ≥ 3.5 cm, a fundus/cervix ratio of > 1, an ovarian volume of ≥ 2 ml, and the presence of an endometrial echo.
To identify any hypothalamic or pituitary lesions, all the girls diagnosed with CPP underwent MRI (Siemens MAGNETOM Symphony 1.5 T MRI), using gadolinium-enhanced T1- and T2-weighted MRI to detect occult intracranial lesions. The standardised MR protocol consisted of 3-mm sagittal and coronal T1-weighted spin-echo slices (TR = 400 ms, TE = 115 ms, matrix size = 512 × 512) centred on the pituitary region, 3-mm axial T2-weighted fast spin-echo slices (TR = 5,040 ms, TE = 115 ms, matrix size = 512 × 512) of the entire head, and 1-mm sagittal T2-weighted 3D gradient-echo slices centred on the midline. The standardised MR protocol consisted of 3-mm sagittal and coronal T1-weighted spin-echo slices (TR = 400 ms, TE = 115 ms, matrix size = 512 × 512) centred on the pituitary region, 3-mm axial T2-weighted fast spin-echo slices (TR = 5,040 ms, TE = 115 ms, matrix size = 512 × 512) of the entire head, and 1-mm sagittal T2-weighted 3D gradient-echo slices centred on the midline.
Ethics approval and consent to participate
All research was carried out in accordance with established Ethical Standards involving human participants. The study was approved by the “Comitato Etico Area di Pavia”, the Ethics Committee of the IRCCS Foundation San Matteo Hospital, on 17 May 2016, reference number 20160005680. All participants provided parental written informed consent to use of the data.
Results were expressed as mean ± SD using the SPSS for Windows version 21.0 (IBM SPSS, Inc., Chicago, IL, USA). A comparison between the two groups (girls with and without pineal cysts) was performed by Mann–Whitney test. Any association between the variables were found using Spearman correlations. Levels of statistical significance were set at p < 0.05.