This study aimed to comprehensively report the epidemiological and clinical features of atypical anorexia nervosa (AAN) in children and adolescents. In May 2024, a systematic review was performed using Medline, Cochrane Library, ClinicalTrials.gov, and relevant websites. Following PRISMA guidelines, 234 articles were screened for studies on DSM-5-defined AAN. A standardized checklist-the JBI critical appraisal tool-was adopted in assessing methodology, and 13 retained studies passed the screening and critical appraisal process for the final review. The Newcastle-Ottawa Scale was utilized to assess the risk of bias in cohort and case-control studies, ensuring a comprehensive evaluation of methodological quality. AAN prevalence in young age groups is 2.8%, with a cumulative 2.8% incidence over 8 years. Incidence is 366 per 100,000 person-years, and the average episode duration is 11.6 months, with a 71% remission rate. Diagnostic persistence for AAN is less stable than other restrictive feeding and eating disorders (FEDs). AAN individuals exhibit higher EDE-Q scores, more severe distress, and distinct BMI differences compared to those with anorexia nervosa and controls. The diagnostic transition from the DSM-IV to the DSM-5 shows that AAN patients are predominantly female, slightly older, and with higher weight. This study yields concrete insights into the features of AAN in the developmental age, highlighting demographic variations, clinical presentations, and treatment outcomes. Recognizing the unique challenges faced by AAN individuals is vital for tailoring effective interventions and improving overall care within the FED spectrum.