Abstract

Materials and Methods

Patients and Study Design

We retrospectively collected data of boys with a diagnosis of CPP between 1 January 1, 2001, and December 31, 2020, from 15 centers of pediatric endocrinology (6 centers in the North, 4 in the Center, and 5 in the South of Italy) belonging to the Italian Society of Paediatric Endocrinology and Diabetology (ISPED).

These patients were selected with specific and homogeneous criteria, as they were in good health, with no neurological signs or symptoms and were referred only for clinical signs of early pubertal development. The diagnosis of CPP was confirmed at each center based on the clinical guideline for an increase in the testicular volume greater than or equal to 4 mL (gonadarche) before the age of 9 years, with an increased height velocity, an advancement of bone age (BA) ≥ 1 year compared to chronological age (CA), a baseline luteinizing hormone (LH) > 0.3 IU/L, and/or LH > 5 IU/L after a GnRH stimulation test (1, 18, 20, 21).

Inclusion criteria were (i) the onset of gonadarche before 9 years of age in otherwise normal healthy boys; (ii) the execution of brain MRI with a detailed examination of the hypothalamic-pituitary area at diagnosis; and (iii) in subjects who underwent GnRH treatment, good treatment compliance documented by clinical and hormonal parameters as well as the BA assessment.

Exclusion criteria were (i) chronic medical conditions already under follow-up at the centers participating in the study and known to be associated with CPP (also including CNS lesions and/or syndromes); (ii) congenital adrenal hyperplasia; (iii) other forms of PPP and isolated premature adrenarche (defined as premature pubic/axillary hair development, normal growth rate, and normal BA after exclusion of serious adrenal disorders) (6).

Figure 1 shows the flowchart of participant enrollment in the study.

Open in a separate window

Figure 1.

Design of the study and overview of recruitment of male patients with CPP.

Patients were divided into 3 groups based on the MRI findings:

1. Group 1, no CNS abnormalities or minor intracranial alterations not involving the hypothalamic-pituitary region

2. Group 2, mild abnormalities of the hypothalamic-pituitary region, considered incidental findings unrelated to CPP; incidental findings were defined as lesions of the pituitary region discovered fortuitously by imaging for reasons unrelated to pituitary disease (22)

3. Group 3, pathological CNS abnormalities known to be associated with CPP

Since there are rather controversial data in the literature (3, 20, 22, 23), the patients were allocated into the different groups through a process of collaboration between clinicians and neuroradiologists. The retrospective nature of the study did not allow us to centralize the MRI reports, but an expert pediatric neuroradiologist was available at each center. In selected cases, the multidisciplinary approach was extended to the pediatric oncologist. In all cases with pathological CNS abnormalities, in the majority of cases with incidental findings (excluding empty sella), and in patients with pineal cysts, a neuroradiological follow-up was performed (2 or more MRI scans during the follow-up).

To evaluate the possible increase in the frequency of CPP diagnosis over the years, all cases were further divided into 4 groups according to the following referral periods: 2001–2005, 2006–2010, 2011–2015, and 2016–2020. Within each referral period the boys were subdivided based on age at diagnosis < or ≥ 7 years

We analyzed 5 groups of variables related to demographics, the visit before GnRH analogue therapy, characteristics of therapy, the visit after 1 year of therapy, and at the end of therapy.

Results

Figure 2 shows the frequency and the etiology of CPP according to the referral periods. The number of cases increased during the last 2 decades and most of these cases were found to be idiopathic. Considering the trend of observed and estimated change over time during the study period, the estimated percent change between the start and end time points was 30% (95CL, 2.9%-10%). The 95% CI shows that the percent change is not statistically significant (Supplementary Fig. S1) (30). An increase in CPP diagnoses was observed only in Group 1 patients with age at referral ≥ 7 years (2001–2005, 8.9%; 2006–2010, 20.9%; 2011–2015, 27.5%; 2016–2020, 30.5%). The frequency of diagnoses remained stable over time in all groups of patients with age at referral < 7 years and in patients in groups 2 and 3 with age at referral ≥ 7 years.

Open in a separate window

Figure 2.

Frequency and etiology of CPP cases divided according to the referral periods: 2001–2005, 2006–2010, 2011–2015, and 2016–2020.

Clinical and Endocrine Features During Follow-Up

Table 4 shows height SDS, BMI SDS, and testicular volume in the 3 groups at diagnosis, after 1 year of therapy, and at the end of therapy. The parameters at diagnosis in the subjects who did not complete follow-up and in the subjects that completed the follow-up were not significantly different (Supplementary Table S1) (30).

Table 4.

Height, BMI, and testicular volume in the 3 groups of male patients with CPP at diagnosis, after 1 year of therapy, and at the end of therapy

	Height (SDS)						BMI (SDS)						Testicular volume (mL)
	N°	At diagnosis	N°	After 1 yr of therapy	N°	End of therapy	N°	At diagnosis	N°	After 1 yr of therapy	N°	End of therapy	N°	At diagnosis	N°	After 1 yr of therapy	N°	End of therapy
Group 1	158	0.8 ± 1.2	126	0.8 ± 1.5	96	0.6 ± 1.5	158	0.7 ± 1.9	122	0.7 ± 1.7	95	0.6 ± 2.1	167	6.9 ± 3.1	133	6.1 ± 1.4	100	6.8 ± 3.5
Group 2	12	0.9 ± 1.2	8	1 ± 0.9	6	1 ± 0.6	12	0.9 ± 0.8	8	1.1 ± 1.1	6	0.7 ± 1.1	15	7.4 ± 3.4	8	8.3 ± 1.4	6	9.1 ± 2.5
Group 3	10	1.3 ± 1.7	6	1.2 ± 1.3	4	1.3 ± 0.3	10	1.3 ± 1.1	6	1.5 ± 0.9	4	1.8 ± 0.5	11	8.6 ± 3.8	6	8.1 ± 9.2	4	9.5 ± 5.7

Open in a separate window

Figure 3 shows the final heights (FH) with the respective target heights (TH) in the 3 groups of patients undergoing GnRH analogue therapy and in 9 patients of the group in which the therapy was not started. The FH appeared to be in the range of the TH in all groups. None of the subjects who did not undergo therapy achieved a FH ≥ TH. The mean TH in the subjects with available FH and in the subjects without available FH was not significantly different (Supplementary Table S2) (30).

Open in a separate window

Figure 3.

Final heights and target heights in the 3 groups of subjects diagnosed with CPP who underwent GnRH analogue therapy and in a group of subjects diagnosed with CPP who did not start therapy.

Discussion

In our large study population of otherwise normal healthy boys referred for PP signs, the percentage of organic forms due to causative brain lesions was 5.7%, one of the lowest reported in the literature. In accordance with all the previous studies, our patients with organic forms showed a significantly lower CA at diagnosis than those with idiopathic forms. Therefore, we observed an increase over time in the number of diagnoses of CPP, particularly with regard to the idiopathic forms.

Compared with historical studies (7, 8), more recent studies published in the last 2 decades already report a lower but more variable prevalence of organic forms (10-16). This variability is attributable to many factors, such as different study design, the low number of cases linked to the monocentric character, and, above all, the different criteria for selecting the patients and the methods of defining the incidental findings.

Our results were consistent with those obtained in a Chinese and a Korean study in which selection criteria very similar to ours were applied (12, 15). Wang et al (15) reported that 16.3% of their cases had sellar abnormalities, of which only 1 was pathogenic, and Yoon et al (12) reported that 7% of the CPP boys with MRI findings were all classifiable as incidental findings. Considering the same results obtained in populations of different ethnicities, we can probably rule out the influence of environmental factors and suggest that this picture should be typical of male CPP itself.

In agreement with Wang et al (15), syndromic forms or medical conditions with psychomotor delay, chromosomal abnormalities, and/or autism were also excluded; according to a recent study, they are more frequently associated with CPP and seem to be more commonly represented in males (32).

In other studies (9, 10, 13, 14), the enrollment of patients already being followed for CNS lesions and/or syndromes such as neurofibromatosis typically associated with CPP introduces an evident selection bias by increasing the overall number of organic forms with CNS abnormalities.

The spectrum of MR lesions reported in the literature is wide and their classification into CNS abnormalities related to CPP, incidental findings or normal variants is uncertain (3, 11, 12, 14, 15, 20, 22, 33-35). In particular, the clinical significance of pituitary microadenoma appears controversial. In fact, considering that it is a histological diagnosis, its relationship with the MR image may be uncertain. Our inclusion of pituitary microadenoma in the group of incidental findings took into account the similarity of clinical and hormonal behavior to that of subjects with normal MRI findings and neuroradiological follow-up. Its inclusion is also in agreement with the conclusions of recent studies on pediatric populations selected with criteria similar to ours (20, 23). As regards the isolated and asymptomatic forms of pineal cysts, these are included by some authors in the normal variants (20, 34) and by others in the incidental findings (22, 29). According to large MR studies the reported prevalence in the population of normal young adults is 2% to 2.5%, similar to that found in our study population (31). Empty sella is classified as incidental findings in all recent studies concerning pediatric patients with CPP (20, 23, 32). Its reported prevalence in the general population is between 8% and 35%, while among adults, the frequency of empty sella found during routine autopsy varies between 5.5% and 12% (22). Finally, both pineal cyst and empty sella have been found in other pediatric conditions (ie, idiopathic short stature) suggesting that they are not specific for CPP (29, 36)

In accordance with all the previous studies, our patients with CNS pathological findings show a significantly lower CA at diagnosis than those with idiopathic forms. In studies that do not use our selection criteria, it is hypothesized that this aspect depends on the follow-up of patients already followed for syndromic forms of CPP (8-11). The presence of similar results in our study and other studies that use the same selection criteria seems to demonstrate that the phenomenon is an intrinsic characteristic of organic forms with associated CNS abnormalities. Unfortunately, from a diagnostic point of view, this parameter does not show a sensitivity of 100%, and 6/11 of our cases with causative brain lesions showed a CA at diagnosis greater than 7 years. Therefore, in the current state of knowledge, while emphasizing the importance of early CA at diagnosis as a discriminating criterion for performing an MRI, it cannot be applied to all cases and thus limits the number of investigations to be performed in male patients with CPP.

The increase over time in the number of diagnoses of CPP, with particular regard to the idiopathic forms, is a finding common to many studies (6, 11-15). In particular, in the Danish study that took into consideration the data from a population registry, a 15-fold increase in CPP incidence in boys with Danish origin (from 0.1 per 10 000 to 2.1 per 10 000) was reported (6) Among the possible influencing factors, better awareness of CPP among local pediatricians, better access to specialized care, and/or parents’ increased knowledge obtained through media are taken into consideration from time to time. The increase over time found in our study only in patients with idiopathic “early” CPP with age at referral ≥ 7 years could lead us to hypothesize the influence of a better awareness among the local pediatricians of these less severe forms of CPP. However, only in an American study is the comparison with a critical potential confounder reported, that is, the overall patient volume which increased in the same referral periods. Therefore, the authors conclude that the increase is only apparent and not real (13). Our data on the trend of observed and estimated change over time appear to be in line with the conclusions of this study.

Other factors, such as the effects of endocrine-disruptor chemicals, improved nutritional supplements, and higher adiposity, are also hypothesized to explain the increase in diagnoses. The Copenhagen Puberty Study, in a population of 1528 Danish boys aged 5.8 to 19.9 years, reported the effect of increased adiposity on the earlier onset of puberty (19). Other studies report a higher BMI in males with idiopathic forms of CPP (11, 14, 15). The results of our study were consistent with the results of Topor et al (13) and do not seem to confirm these data. In fact, both at diagnosis and during follow-up in Groups 1 and 2, most patients show BMI values within +2 SD. Only in subjects with organic forms, we observed higher average BMI values, which could, however, be related to the organic pathology itself and might not represent a characteristic of CPP in males.

The presence of a positive family history of PP only in boys with idiopathic CPP confirms the influence of genetic factors reported in various recent studies, especially in boys with CPP (3, 37-40). The search for genetic variants is beyond the scope of the current study, but our results confirm the usefulness of performing a genetic analysis before proceeding with further diagnostic tests (3). In fact, Bessa et al demonstrated a high frequency of MKRN3 mutations in a cohort of 20 boys with CPP, previously classified as idiopathic, suggesting the importance of genetic analysis in this group (38).

Our results regarding follow-up and therapeutic outcome have limited value, in particular because prolonged follow-up is available in few subjects, especially in groups 2 and 3. On the other hand, there is a scarce amount of data in the literature regarding a cohort of boys with CPP selected with homogeneous criteria similar to ours. In accordance with the results of the study by Mul et al (41), our patients reach, on average, a FH that is not significantly different from the TH, but the results of the 2 studies are only partially comparable due to their different study design and enrollment of patients. The strengths of our study are its considerable size—the largest national series of Italian boys with CPP examined in the last 2 decades—and the homogeneous diagnostic approach used in the pediatric endocrinology centers participating in the study.

The limitations are the retrospective nature of the study, the unavailability of the total number of boys referred and evaluated over the same period, including those who did not meet criteria for CPP, and the non-centralization of the MRI and BA reports. Radiological investigations were in all cases evaluated only by expert pediatric personnel.

In conclusion, in the data we obtained from a large nationwide cohort of otherwise normal healthy boys referred for PP signs, the percentage of organic forms due to causative brain lesions is lower compared with previous literature data. Based on current knowledge, it is still not possible to have a safety marker predictive of organic forms of CPP that would establish a limit for MRI examination in the idiopathic ones.

Abbreviations

Contributor Information

Alessandra Cassio, Pediatric Unit, IRCCS Azienda Ospedaliero-Universitaria di Bologna, 40138 Bologna, Italy. Department of Medical and Surgery Sciences, University of Bologna, 40138 Bologna, Italy.

Gloria Marescotti, Pediatric Unit, IRCCS Azienda Ospedaliero-Universitaria di Bologna, 40138 Bologna, Italy.

Tommaso Aversa, Department of Human Pathology of Adulthood and Childhood, University of Messina, 98125 Messina, Italy. Pediatric Unit, University Hospital, 98122 Messina, Italy.

Mariacarolina Salerno, Pediatric Endocrinology Unit, Department of Translational Medical Sciences, University Federico II, 80131 Naples, Italy.

Gianluca Tornese, Institute for Maternal and Child Health IRCCS “Burlo Garofolo”, 34137 Trieste, Italy. Department of Medicine, Surgery and Health Sciences, University of Trieste, 34129 Trieste, Italy.

Marianna Stancampiano, Department of Pediatrics, Endocrine Unit, IRCCS San Raffaele Scientific Institute, 20132 Milan, Italy.

Gerdi Tuli, Department of Pediatric Endocrinology, Regina Margherita Children's Hospital, Department of Pediatric Sciences, University of Turin, 10126 Torino, Italy.

Maria Felicia Faienza, Pediatric Unit, Department of Precision and Regenerative Medicine and Ionian Area, University of Bari “A. Moro”, 70124 Bari, Italy.

Paolo Cavarzere, Pediatric Division, Department of Pediatrics, University Hospital of Verona, 37126 Verona, Italy.

Daniela Fava, Department of Pediatrics, IRCCS Istituto Giannina Gaslini, Genoa 16147, Italy. Department of Neuroscience, Rehabilitation, Ophthalmology, Genetics, Maternal and Child Health, University of Genoa, Genoa 16142, Italy.

Maria Parpagnoli, Department of Diabetology and Endocrinology, Meyer Children Hospital IRCCS, 50139 Florence, Italy.

Patrizia Bruzzi, Pediatric Unit Department of Medical and Surgical Sciences of Mothers, Children and Adults, University of Modena and Reggio Emilia, 41124 Modena, Italy.

Anastasia Ibba, Pediatric Endocrinology Unit and Newborn Screening Center, Pediatric Microcitemic Hospital, ASL Cagliari, 09121 Cagliari, Italy.

Valeria Calcaterra, Child and Adolescent Unit, Department of internal Medicine and Therapeutics, University of Pavia, 27100 Pavia, Italy,

Chiara Mameli, Department of Biomedical and Clinical Sciences, University of Milan, 20157 Milan, Italy.

Anna Grandone, Department of Woman, Child and of General and Specialized Surgery, Università degli Studi della Campania “Luigi Vanvitelli”, 80138, Napoli, Italy.

Valentino Cherubini, Department of Women's and Children's Health, Salesi Hospital, 60126 Ancona, Italy.

Valentina Assirelli, Pediatric Unit, IRCCS Azienda Ospedaliero-Universitaria di Bologna, 40138 Bologna, Italy. Department of Medical and Surgery Sciences, University of Bologna, 40138 Bologna, Italy.

Francesca Franchina, Department of Human Pathology of Adulthood and Childhood, University of Messina, 98125 Messina, Italy. Pediatric Unit, University Hospital, 98122 Messina, Italy.

Donatella Capalbo, Pediatric Endocrinology Unit, Department of Mother and Child University Hospital Federico II Naples, 80131 Naples, Italy.

Raffaella Di Mase, Pediatric Endocrinology Unit, Department of Mother and Child University Hospital Federico II Naples, 80131 Naples, Italy.

Gianluca Tamaro, Institute for Maternal and Child Health IRCCS “Burlo Garofolo”, 34137 Trieste, Italy.

Julia Cavasin, Department of Medicine, Surgery and Health Sciences, University of Trieste, 34129 Trieste, Italy.

Jessica Munarin, Department of Pediatric Endocrinology, Regina Margherita Children's Hospital, Department of Pediatric Sciences, University of Turin, 10126 Torino, Italy.

Gianni Russo, Department of Pediatrics, Endocrine Unit, IRCCS San Raffaele Scientific Institute, 20132 Milan, Italy.

Malgorzata Wasniewska, Department of Human Pathology of Adulthood and Childhood, University of Messina, 98125 Messina, Italy. Pediatric Unit, University Hospital, 98122 Messina, Italy.

the Physiopathology of Growth Processes and Puberty Study Group of the Italian Society for Pediatric Endocrinology and Diabetology:

References