Increased incidence of precocious and accelerated puberty in females during and after the Italian lockdown for the coronavirus 2019 (COVID-19) pandemic

Puberty is the result of a complex neuroendocrine system, initiated by an unknown primary mechanism, and characterized by the increased release of gonadotropin-releasing hormone (GnRH) by the hypothalamus [1, 2]. The phenomenon is related to a complex sequence of endocrine changes and is influenced by integrated multiple peripheral and central signals leading to the development of sex characteristics, the growth spurt and the obtaining of reproductive competence [1, 2].

Currently, abnormally precocious sexual development (precocious puberty or PP) is defined as the occurrence of Tanner stage B2 before 8 years in girls and Tanner stage G2 before 9 years in boys. The average age of onset of puberty has decreased over the last century [3‐5]. Historical data from Europe show a sharp decline in mean age at menarche from approximately 17 years in the early nineteenth century to approximately 13 years by the mid-twentieth century, with a minor decrement during the past 25–30 years [5]. Mean age at onset of breast development has also decreased in all ethnic groups [6], as reported by the Pediatric Research in Office Settings (PROS) study [3] and the data by Biro et al. [7]. These same data show a progressive increase in the number of girls fulfilling the criteria for PP (6.7% in the PROS study and 10.4% in the Biro data for white girls) [6, 7].

Genetic factors undoubtedly play a major role in these trends [8]. However, there is a consensus that environmental variables, such as weight, foetal nutrition, childhood dietary habits, physical activity, psychological factors, and exposure to electromagnetic fields (EMF) and/or endocrine disrupting chemicals, also have an impact [2, 5, 9].

In Italy, as elsewhere, the COVID-19 pandemic created an unprecedented strain on health care and during the strictest months of lockdown forced millions of people to remain isolated in their homes for about 2 months, from March to May 2020 [10, 11]. It is possible that the lockdown enhanced the impact of factors interfering with the timing and tempo of puberty, such as adiposity [12, 13] and psychological wellbeing. An increased incidence of depression, anxiety and stress have been reported to varying degrees in adults [14], and even if comprehensive data among children and adolescents are scarce, it is likely that children had similar psychological reactions.

The aims of this study are to evaluate the frequency of precocious puberty and the rate of pubertal progression in girls during and after the lockdown, comparing results with data for the same period over the last 5 years and to consider possible correlations between lockdown conditions and trends in the data.

Patients with a diagnosis of CPP and associated hypothalamic–pituitary congenital malformations, neurological, neurosurgical and/or genetic diseases, psychomotor delay, oncological diseases, other endocrine impairments requiring hormonal treatments, or taking drugs that may interfere with pubertal development were excluded from the study. We also excluded children who were adopted or had immigrated to Italy as such children have a statistically higher rate of precocious puberty than the general pediatric population.

The study was performed according to the Helsinki II declaration and approved by the local Paediatric Ethical Committee (approval number: 30/07/2020–179). Written informed consent was obtained from the parents of enrolled children.

Clinical data and laboratory results were analysed in 49 females. All patients showed normal values of 17-OHP, TSH and FT4. Contrast enhanced brain MRI showed a normal hypophysis without focal abnormalities in all patients.

Thirty-seven patients (Group 1) received a new diagnosis of CPP during and after the Italian lockdown and 12 (Group 2), already diagnosed with slow progression CPP, presented an unexpected accelerated rate of pubertal progression in this period (Tables 1 and 2). The data of the new cases of CPP were compared with 89 patients with CPP diagnosed in the same period (March – July) of the years 2015–2019 (17.8 ± 1.3 patients, range from 16 to 19 cases/same period/year). The data of the 12 patients with slow progressive CPP in whom pubertal progression accelerated were compared with 11 patients with the same characteristics who had been followed by our Unit in the previous 5 years (2.2 ± 0.4 patients, ranging from 2 to 3 cases/same period/year).

Four patients were excluded from the study: two for CPP and cognitive delay (one of whom was adopted), one for a hypothalamic-pituitary malformation and one for puberty associated with Williams-Beuren syndrome.

The overall number of new CPP diagnoses was significantly higher during and after lockdown for COVID-19 than the mean number of cases for the same period each year from 2015 to 2019 (p < 0.0005, Table 1 and Fig. 1). The number of patients with slow progression CPP who transitioned to fast progression CPP during and after the lockdown was also significantly higher (p < 0.0005, Table 2 and Fig. 2).

Compared with control group 1, the girls in group 1 of this study presented an earlier chronological age at B2 (p < 0.005), an earlier chronological age at diagnosis (p < 0.0005), and a more advanced Tanner stage at diagnosis (p < 0.05). However, there was no significant difference between the time between the appearance of the B2 and diagnosis in group 1 patients and controls (respectively 3.1 ± 0.9 vs. 3.0 ± 0.8 months).

The patients in group 1 also had more markedly elevated basal LH and E2 levels and a higher LH peak after LHRH test in comparison with the control group (p < 0.005). They also presented greater increases in uterine length and ovarian volume (p < 0.0005). Auxological parameters such as height SDS, BMI SDS and bone age SDS were not statistically different between group 1 patients and controls, but BMI SDS increased significantly in group 1 during the lockdown (p < 0.05).

Compared with the previous 5 years, during lockdown, the number of patients with CCP who experienced a transition from slowly progressive pubertal development to accelerated pubertal development was statistically significant (p < 0.0005). Compared with controls, patients in group 2 at the same stage in follow-up (0.31 ± 0.02 vs. 0.38 ± 0.02 years), had significantly elevated LH (p < 0.05) and E2 levels (p < 0.005), as well as a higher LH peak after LHRH test (p < 0.05) and a significantly accelerated progression rate as measured by the Tanner scale (p < 0.0005), uterine length (p < 0.05), and ovarian volume (p < 0.05). As in group 1, for patients in group 2, auxological parameters (height SDS, bone age SDS) were not statistically different from controls but Δ BMI SDS increased significantly (p < 0.005).

The evaluation of the use of electronic devices before and during the lockdown in the two groups, showed a significantly increased use (before: 1.6 ± 0.9 h/day; during: 3.9 ± 1.5 h/day, p < 0.0005). The main use was for school lessons and study (1.8 ± 0.8 h/day), but also for entertainment (TV and videogames) (1.3 ± 0.8 h/day vs. 0.7 ± 0.5 h/day, p < 0.0005). Use of devices during the hours before sleeping also significantly increased (p < 0.0005).

Our study shows a significantly increased incidence of new diagnoses of central precocious puberty and a faster rate of pubertal progression in previously diagnosed patients during and after the lockdown for COVID-19 in comparison with the same period of the 5 previous years. Our results seem to suggest a correlation between environmental factors and early onset and fast progression puberty. Our study focuses on female patients, but we hypothesize that such factors may also affect pubertal development in boys. As diagnosis in males is more frequent at the G2 stage of genital development, it would be interesting to evaluate any increases in the frequency of CPP in male patients over the coming months.

In humans, 70–80% of variances in pubertal timing and tempo can be explained by genetic factors, but the role of environmental factors is also well known [2, 5]. A number of studies, reported an earlier menarcheal age in girls with a low birth weight or in utero growth retardation, showing that the possible effects of the environment start during intrauterine life [26], although other studies contradict this conclusion [27].

Our study may help shed light on 3 factors hypothesized in the literature as potentially contributing to the timing and/or tempo of pubertal development: an increased BMI [28, 29], the “overuse” of electronic devices [21] and psychological triggers [30].

Nutrition plays a key role in the timing of puberty and could explain, at least partially, the secular trend of earlier development [31]. Obesity is associated with early menarcheal age [29, 32]. Leptin, insulin-like growth factor-I and glucose have been shown to be involved in the control of GnRH secretion, but their role in the timing of puberty remains controversial [31]. The effects of fat mass may also interact with the effect of some endocrine disrupters that have a high affinity for lipids and are stored in fat tissue, thus creating conditions for the persistence of systemic effects [31].

Psychological factors may also be important. It is likely that the COVID-19 pandemic impacted the mental health and well-being of children considerably. Illness, anxiety about becoming ill as well as prolonged social distancing may have lasting effects on children [33]. The closure of schools, the sudden disruption of extended social and family relationships, the change in daily habits and parental anxiety about financial and other problems may have affected children’s emotional stability and sense of security [33]. As yet there is little data on how children in all countries responded psychologically to the crisis but [34] the importance of psychological factors in precocious puberty has been reported in children migrating from developing to developed countries, primarily through international adoption [35]. It has been hypothesised that for adopted children who move from a deprived environment to a resource rich environment, environmental factors which slow or delay pubertal development are deactivated while factors which favour pubertal development are enhanced, triggering the onset of puberty [35]. Genetic factors cannot fully explain the high incidence of precocious puberty in internationally adopted girls as their average menarcheal age is lower than the average menarcheal age both in their foster countries and in their countries of origin [35].

Our data show an increased use of electronic devices; the time spent by the children on electronic devices increased by a factor of 2.5 during lockdown. We may hypothesize that the overuse of electronic devices contributes to the timing and/or tempo of puberty. A number of studies have recently investigated the effects of exposure to electromagnetic fields on melatonin [21, 36‐38]. Exposure to simulated electromagnetic fields has been associated with decreased melatonin production by isolated pinealectocytes in vitro, as well as with a decreased pineal and plasma melatonin and its urinary metabolites [36]. The study in the town of Cavriglia showed an overall and age-dependent decrease in urinary melatonin concentration associated with children’s exposure to a television screen which was more pronounced in younger children [21]. Night-time serum melatonin levels are highest in very young children and drop progressively by 80% throughout childhood until adolescence and young adulthood [39]. There is some evidence that the drop in nocturnal melatonin levels parallels sexual maturation processes [40], as measured by the Tanner scale [41]. Animal studies also show that a reduction in melatonin may accelerate pubertal development [42] and that exogenous melatonin can suppress GnRH secretion [43]. It is plausible that a greater use of electronic devices leads to a reduction in melatonin levels which in turn triggers the endocrine changes that lead to the onset of pubertal development at an earlier age [44].

We are aware that this study has several limitations, including its small sample size. The COVID-19 pandemic caused profound and sudden changes in community care and hospital settings, making it impossible to create a homogeneous control group. We underline the need for more studies involving larger numbers of CPP patients. Moreover, our study did not assess decreases the changes in physical activity or in calorie intake in these children. However, increases in calorie intake and decreases in physical activity have been observed in the Italian population during lockdown [13] and it is likely that also our children were affected in a similar way.

In conclusion, our data shows an increase in the incidence of new CPP diagnoses, as well as a faster rate of pubertal progression in previously diagnosed patients, during and after lockdown for COVID-19, suggesting that environmental factors, such as the BMI and the use of electronic devices, which emerged during the lockdown played a triggering role. Further studies are needed to confirm our data and elucidate the relative importance of the various factors at play and how they interact. In this uncertain period, clinicians should be aware of the signs of pubertal developmental disorders and should carefully monitor existing patients with CPP for signs of accelerated pubertal progression.