COVID-19 lockdown and the rate of central precocious puberty

Precocious puberty (PP) is defined as the onset of pubertal development at an age 2–2.5 standard deviations below the average age of normal puberty, which corresponds to 8 years in females and 9 years in males [1, 2]. The onset of normal pubertal development, however, varies significantly according to several factors, such as children’s ethnicity or body weight [3]. Moreover, several studies have shown that nowadays thelarche in girls appears earlier than it did in the last century [2, 4, 5], while historical data collected in the USA and Europe have shown a significant anticipation of the age of menarche, from about 17 years in the early nineteenth century to about 13 years in the mid-twentieth century [6]. This evidence therefore highlights the role of the environment (as in better socioeconomic conditions, endocrine disruptors, etc.…) in influencing the timing of pubertal onset in children.

At the end of 2019, a new coronavirus—named SARS-CoV-2—was identified in Wuhan, a city in the Chinese province of Hubei, and was responsible for a major wave of new cases of interstitial pneumonia that was named COVID-19 [7, 8]. The virus spread rapidly, causing a global pandemic which led to up to 500 million confirmed diagnoses of COVID-19 [8]. Due to the high rate of spreading of COVID-19, several governments were forced—if reluctantly—to impose restrictive policies promoting social isolation and home confinement to contain the continuous increase of pneumonia cases. The Italian government imposed the national lockdown on March 9th, 2020: as of that moment, the population was compelled to stay at home. Among the consequences of this restrictive measure, it is important to point out a drastic change in children’s lifestyle, which lasted several months [9].

Several Italian and international studies have observed that the incidence of new cases of Central Precocious Puberty (CPP) increased significantly during this period of restrictions, especially in female subjects [9‐15]; moreover, the rate of pubertal progression in patients already suffering from CPP also increased compared with the rate observed in the previous years [10]. Despite several hypotheses regarding the mechanism responsible for such an increased incidence of CPP cases following the 2020 lockdown have been put forward, the real pathogenic explanation has yet to be uncovered.

In the present retrospective observational study we therefore meant to compare the incidence of idiopathic central precocious puberty in a single Tertiary Endocrinological Center before and after the onset of COVID-19 lockdown; we also aimed to identify potential difference in anamnestic, clinical, biochemical and/or radiological characteristics between affected girls from the two time periods.

According to our inclusion and exclusion criteria, out of 130 children who were referred to our Center for suspected precocious puberty between 2014 and 2021, 81 patients were ruled out for reasons explained in Fig. 1. Thus, a total of 49 girls with CPP were ultimately included in the data analysis and divided into Post-lockdown Group (19 patients, mean age at CPP onset 7.31 years) and Pre-lockdown Group (30 patients, mean age at CPP onset 6.9 years). Post-lockdown Group included 4 foreign-born subjects and 2 girls who were adopted; Pre-lockdown Group, instead, included 3 foreign-born patients, 4 adopted girls, and 1 girl who spent part of her childhood in a country other than Italy (where she had different eating habits).

The comparison between anamnestic, clinical, biochemical, and imaging characteristics of patients in whom precocious puberty started before (Pre-lockdown Group) and after (Post-lockdown Group) the beginning of 2020 lockdown is shown in Supplementary Tables 2–8.

Like other Italian and foreign Centers previously did, we also have registered an increase in the incidence of idiopathic CPP: to be specific, an almost threefold increase from March 2020 to July 2021 compared to the pre-lockdown period (from 2014 to February 2020).

At present, there is no certainty about the pathogenic mechanism that was able to trigger such an important increase in the incidence of precocious puberty across different Countries. However, it is speculated that the lockdown itself may have exacerbated the influence of certain known environmental risk factors able to promote the onset of precocious puberty in children, such as weight gain [10, 12, 16], psychological distress [17], an increased exposure to particular endocrine disruptors (such as triclosan, which is contained in soaps and hand sanitizers [15, 18, 19]) and a longer time spent using electronic devices [10, 11].

Those mentioned above are theories about what may have contributed to trigger the significant increase in the incidence of CPP as of the beginning of 2020. It is still not clear which one or which ones of these possible explanations, if any, were actually involved, but most certainly something must have happened during the lockdown that interfered with the normal genetically determined timing of pubertal onset of these children. Consistently with this statement, our study revealed that menarche occurred at a significantly older age in mothers of post-lockdown patients compared to those of pre-lockdown patients (p value 0.0039). This certainly constitutes a very interesting finding: indeed, it is well known that the age of pubertal onset is determined by a complex interplay of multiple environmental and genetic factors [20], with genetic factors estimated to be responsible for approximately 50–75% of the variability in the age of onset of the first pubertal signs [21]. A few years ago, a genome-wide association study (GWAS) showed a substantial overlap in genes possibly influencing pubertal timing in both males and females [22], indirectly suggesting that the timing of maternal puberty may in turn be related to the pubertal timing of both sons and daughters [23]. In addition, multiple studies have shown that the age at menarche of mothers is associated with their daughters’ age at menarche as well [23], and a few studies have also assessed a link between maternal age at menarche and the age at each of their either male or female children’s pubertal developmental milestones. For example, a study by Sorensen et al. [23], conducted on more than 15,000 children, showed an association between maternal age at menarche and the age at all pubertal markers in both sons and daughters, a result that appears consistent with the genome-wide association study abovementioned [22]. Similarly, a study by Wohlfahrt-Veje et al. [24] observed that mothers’ age at menarche was associated with both the time of testicular volume increase and pubarche in sons and the time of thelarche in daughters. Based on this evidence, given the higher age at menarche of mothers of post-lockdown patients, we would have expected that in such group thelarche should have appeared later than in Pre-lockdown patients; on the contrary, not only did Post-lockdown patients develop a precocious puberty, but their age at the time of the first pubertal signs did not differ compared to Pre-lockdown Group patients’ (p value 0.5449), as if the dissimilarity in maternal age at menarche had no influence at all. This result allows us to hypothesize that the abovementioned postulated environmental risk factors for precocious puberty that were exacerbated by the lockdown may have played a predominant role and a stronger influence than genetics itself in determining the timing of pubertal onset of post-lockdown girls, interfering with what should have been their genetically determined pubertal timing, anticipating its onset. It allows to strengthen the hypothesis that “something” must have happened and impaired the normal onset of puberty, something that goes beyond the mere genetics. Moreover, noting that the age at thelarche in daughters was less strongly associated with parental pubertal timing compared to the age at menarche, the same Wohlfahrt-Veje and colleagues speculated that the age at thelarche might be influenced more by environmental factors rather than genetics [24], which would be in agreement with our hypothesis. One could possibly speculate that the time of paternal puberty might also have influenced the time of pubertal onset of our patients, and that therefore early-developed fathers might have favored a greater precocity of pubertal development in their children, bringing into play an additional variable (about which we have no information in our study). However, again in the study by Wohlfahrt-Veje and colleagues [24] the Authors found that having an early-developed father was not associated with an earlier age at thelarche in daughters, but only with an earlier age at menarche and pubarche. From these results we can therefore hypothesize that paternal influence in the age of thelarche on our female patients was negligible.

Furthermore, in patients who were diagnosed with CPP after the lockdown we found an earlier age at diagnosis and a reduction in the time interval between the appearance of the first pubertal signs and the diagnosis itself compared with patients from previous years: these differences, despite showing only a trend toward statistical significance, are in line with what was observed in the study by de Oliveira Neto et al. [16], in which the time to reach diagnosis was even halved in post-lockdown patients. It is possible that the prolonged period of home isolation gave parents the opportunity to recognize the pubertal changes of their children more promptly; an increased attention to their children’s physical changes was most likely maintained also in the months following the end of lockdown, considering that phenomena such as smart working, e-learning and the persistence of restrictions on social, touristic, sportive and restaurant activities encouraged a further prolongation of the time parents and their children spent together at home. A second explanation, as suggested by de Oliveira Neto et al. [16], could instead be a more rapid progression of pubertal development in post-lockdown CPP patients, which may have made the signs of pubertal development more evident, thus facilitating their discovery by parents and anticipating medical contact. This second hypothesis seems to be in line with what Stagi et al. [10] observed, namely that patients already suffering from CPP experienced an accelerated pubertal progression during and after lockdown. In contrast, in our study both pubic hair and breast development detected at the time of the first clinical evaluation were found to be comparable between the two groups of patients.

Finally, even though the Post-lockdown girls of our cohort presented to medical attention with a bone age advancement by at least 1 year compared to chronological age, and a LH peak after GnRH testing diagnostic for the pubertal range [25], there are several elements that the Authors found suggestive of a less progressive form of CPP compared to Pre-lockdown patients. Firstly, we found that Post-Lockdown girls had a smaller percentage increase in LH (delta LH%) after GnRH testing compared to Pre-lockdown girls. Secondly, Post-lockdown girls had a lower bone age advancement (RUS TW3) compared to Pre-lockdown patients. Lastly, Post-lockdown girls had lower delta4-androstenedione levels compared to Pre-lockdown girls: this was interpreted by the Authors as a lower ovarian activation, since both 17OH-P and DHEA-S levels were normal (therefore ruling out adrenal activation), yet comparable between the two groups (except for one single patient in the Pre-lockdown Group with slightly increased DHEA-S levels). When taken together, all these results lead us to hypothesize that Post-lockdown patients might have undergone a less progressive form of CPP compared to Pre-lockdown patients, although only the long term follow-up of these patients—ideally on a 6–12 months follow-up—will be able to confirm or rule out such hypothesis (taking into account growth-velocity, pubertal development progression, age at menarche, bone age advancement). Hopefully, these observations might be confirmed on multicenter studies.

In conclusion, our data confirmed an almost threefold increase in new diagnoses of CPP as of the beginning of lockdown compared to the previous six years, suggesting that during the pandemic the influence of genetics in determining the timing of pubertal onset has been scaled back in favor of a stronger environmental influence (the lockdown-effect).

Furthermore, despite a trend towards an earlier diagnosis of CPP was registered during the pandemic, post-lockdown patients seem to be characterized by a less progressive form of pubertal advancement compared to pre-lockdown patients, as suggested by a slightly decreased hypothalamus-pituitary–gonadal axis activation, a less advanced bone age and a significantly reduced ovarian activation (lower delta-4-androstenedione levels). These data would need to be confirmed in a longer observational study focused on auxological follow-up of these children, taking into account hormonal levels, bone age advancement, growth velocity and timing of menarche.