Eating disorders are defined as a group of diseases characterized by altered eating behaviors and body image concerns. In adulthood, evidence suggests an association between photo manipulation, body dissatisfaction, and reduced self-esteem. It has also been suggested that photo manipulation could reinforce risks of body shame, depression, and eating disorders [7, 8]. Thus, body image perception represents a key risk factor for the development of eating disorders [9, 10]. Posting edited selfies is common between minors, with more than 90% of the teens that use social media posting photos of themselves Nevertheless, scientific evidence on the consequences of photo manipulation lacks. Examining the literature, a relationship between disordered eating and body image has been found in adolescent girls with a greater effort in choosing a selfie and monitoring likes/comments. So, photo manipulation with digital editing of a selfie, seems to be linked not only to decreased personal perception, but also to eating concerns. Literature revision provide insight in an association between appearance-related social media behaviors and eating disorders in minors [1, 3].

Photo editing exposes females to greater peer scrutiny of appearance and competition, which, from an evolutionary perspective, has been proposed to increase body dissatisfaction and body concerns. Taking, sharing and comparing selfies through social media, may heighten appearance focus and increase internalization of appearance ideals, contributing to body dissatisfaction. In addition, monitoring of likes/comments on the uploaded selfies may represent an online manifestation of reassurance-seeking [11].

New social media and online chat, including Facebook, Instagram, Snapchat, and TikTok exposure may be associated with lower personal weight satisfaction, higher drive for thinness, and eating disorder symptoms. Studies revealed that time spent on the Internet is significantly related to internalization of the thin ideal, body surveillance, and drive for thinness. In particular, Facebook users scored significantly higher on all body image concern measures than non-users [12]. The more vulnerable the subjects are, the greater the risk is. Young adolescent girls with high levels of body-related and eating concerns might engage in social media activities manipulating selfies to present an ideal appearance when sharing images. In the paper by McLean et al., 20 out of 74 photos had been manipulated. In the selfie-sharing group, higher scores for selfie manipulation were associated with higher scores for body-related and eating concerns [1].

A topic of concern is that social media use among adolescents is increasing, despite the fact that social media access for minors is restricted by age in many countries. In Italy, many platforms, including TikTok, Instagram, Facebook, and YouTube, have age restrictions. However, without a strict parental control, minors can easily access social media platform and download the applications.

Even though social media may help adolescents to build social networks, connect, chat and stay in touch with old and new friends, their use may also be linked to potential harms and unhealthy mental effects, following exposure to inappropriate videos and images, especially in vulnerable categories like adolescents.

A limitation of the study is the few available data connected to few evidences on the pediatric population. Further studies may add important information and improve awareness of this phenomena. Even if our systematic has been based by a detailed strategy adopting PRISMA guideline, we have to declare the risk of selection bias as we limited our review to Pubmed database as is the most extensively used database and search engine in the biomedical and healthcare fields, contain more than 37 million citations and abstracts of biomedical literature. We excluded other datasets, unpublished papers, clinical trial registries, regulatory agency websites, and conference abstracts. Another bias may regard contributing reports focusing on the provenience of the examined population, mainly Australian, and on the different inclusion criteria used by the authors which contribute heterogeneity to the results. Finally, as we got just two findings, no sensitivity analyses have been adopted to compare the reports.