Background/Objectives: Skeletal dysplasias are a heterogeneous group of rare genetic disorders with diverse and overlapping clinical presentations, posing diagnostic challenges even for experienced clinicians. With the increasing availability of artificial intelligence (AI) in healthcare, large language models (LLMs) offer a novel opportunity to assist in rare disease diagnostics. This study aimed to compare the diagnostic accuracy of two advanced LLMs, ChatGPT (version GPT-4) and DeepSeek, with that of a clinical expert panel in a cohort of pediatric patients with genetically confirmed skeletal dysplasias. Methods: We designed a prospective vignette-based diagnostic benchmarking study including 45 children with confirmed skeletal dysplasias from two tertiary centers. Both LLMs were prompted to provide primary and differential diagnoses based on standardized clinical case vignettes. Their outputs were compared with those of two human experts (a pediatric endocrinologist and a pediatric orthopedic surgeon), using molecular diagnosis as the gold standard. Results: ChatGPT and DeepSeek achieved a comparable diagnostic top-3 accuracy (62.2% and 64.4%, respectively), with a high intermodel agreement (Cohen’s κ = 0.95). The expert panel outperformed both models (82.2%). While LLMs performed well on more common disorders, they struggled with ultra-rare and multisystemic conditions. In one complex case missed by experts, the DeepSeek model successfully proposed the correct diagnosis. Conclusions: LLMs offer a complementary diagnostic value in skeletal dysplasias, especially in under-resourced medical settings. Their integration as a supportive tool in multidisciplinary diagnostic workflows may enhance early recognition and reduce diagnostic delays in rare disease care.

Children with a classic form of congenital adrenal hyperplasia (CCAH) have a potentially increased risk of unfavorable cardiometabolic events due to the interplay of corticosteroid treatment, hyperandrogenism, and other factors. Although readily recognized in adults, these aspects are frequently overlooked in children and youth with CCAH; Aim: To review the evidence available from studies regarding cardiometabolic health outcomes in CCAH patients; Methods: A review of the literature was performed following PRISMA guidelines, including studies published between 2000 and 2024. We included studies reporting cardiometabolic outcomes in children and adolescents (<18 years) with CCAH. Where pediatric data were sparse, additional data were obtained from studies with older adolescents and young adults (15–25 years). Cardiometabolic outcomes included risk factors, such as obesity, insulin resistance, lipids, blood pressure, and vascular markers; Results: Twenty-five studies were analyzed. The prevalence of obesity was found to be higher in children with CCAH, as well as of increased visceral adiposity. Higher indices of insulin resistance were also a frequent finding in children with CCAH. CCAH patients had higher systolic blood pressure and more frequently loss of nocturnal blood pressure dipping, particularly among salt-wasting subtypes and in younger children. Subclinical atherosclerosis was indicated by increased carotid intima–media thickness, elevated hs-CRP, and impaired endothelial function. Other findings suggested changes in lipid profiles, particularly decreased HDL-c and increased triglycerides, although the findings were less consistent; Conclusions: Compared with the general pediatric population, children with CCAH were found to have an increase in multiple cardiometabolic risk factors. It is therefore vital to monitor these risk factors in pediatric CCAH, as well as tailoring treatment with cardiometabolic health in mind, to achieve better long-term cardiovascular and metabolic outcomes. Future research should focus on longitudinal studies of cardiometabolic outcomes and innovative therapeutic approaches to reduce these risks in patients with CCAH.

Introduction: Aniridia is a rare congenital eye anomaly associated with various ophthalmological problems. Some of the causes of this disease are mutations in the PAX6 gene or deletion of loci on the short arm of chromosome eleven, which can be complete or partial. In this study, PAX6 gene sequence was analysed in four unrelated patients. Aim: Examination of the entire coding sequence of the PAX6 gene in our patients in order to detect mutations in this gene. Material and Methods: Patients' DNA was isolated from samples of peripheral blood. All exons and flanking intronic sequences of the PAX6 gene were amplified using PCR and screened for mutation with direct DNA sequencing by Sanger method. Results: In one patient, the insertion type of mutation (c.342-2insC) was found in the second intron of the PAX6 gene. Other three patients had normal PAX6 sequence although they had phonotypical expression of aniridia. Conclusion: ttese are first results of PAX6 gene analysis in Serbia. Apart from most common mutations, deletions of various extent in coding regions of PAX6 gene, other types of mutations are also possible causes of aniridia. The insertion found in this study has not been reported before. It is also possible that mutations in other genes are associated with iris malformation, which can explain the finding of normal PAX6 gene sequence in three patients with aniridia.