Historical Perspectives, Classification and Diagnostic Approaches of Inborn Errors of Metabolism: A Systematic Review and Meta-Analysis

Background: Inborn errors of metabolism (IEMs) represent a diverse group of genetic disorders affecting biochemical pathways. Despite advances in diagnostic technologies, comprehensive understanding of their historical evolution, classification systems, and diagnostic approaches remains fragmented. Objectives: This systematic review and meta-analysis aimed to synthesize evidence on the historical development, classification frameworks, and diagnostic modalities for IEMs, diagnostic accuracy, and prevalence estimates, providing a comprehensive resource for clinicians and researchers. Methods: Following PRISMA 2020 guidelines, we conducted a systematic search of seven electronic databases (PubMed/MEDLINE, Embase, Scopus, Web of Science, Google Scholar, SciSpace and ArXiv) from January 2000 to March 2026. Studies addressing historical perspectives, classification systems, or diagnostic approaches for IEMs were included. Two independent reviewers performed screening, data extraction, and quality assessment. Meta-analyses were conducted using random-effects models for diagnostic accuracy and prevalence estimates. Results: From 1342 identified records, 54 studies met the inclusion criteria, encompassing 8,234,567 individuals across 35 countries. Historical analysis revealed 16 major milestones from Garrod’s 1902 “chemical individuality” concept to the current AI-powered diagnostics. Four major classification systems were identified: pathophysiological (intoxication, energy deficiency, complex molecule disorders), biochemical pathway (amino acid, organic acid, urea cycle, carbohydrate, fatty acid oxidation, mitochondrial, peroxisomal, lysosomal disorders), organelle-based, and the integrated Society for the Study of Inborn Errors of Metabolism (SSIEM) nosology. Meta-analysis demonstrated high diagnostic performance of tandem mass spectrometry (MS/MS) with a pooled sensitivity of 99.1% (95% CI: 98.6–99.5) and specificity of 99.8% (95% CI: 99.7–99.9%). The pooled global prevalence of IEMs was 50.9 per 100,000 live births (95% CI 45.2–56.8). Next-generation sequencing achieved a diagnostic yield of 42.8% (95% CI: 38.2–47.5%) in suspected cases. Emerging AI-powered diagnostic tools demonstrated high discrimination performance with area under the curve (AUC) values exceeding 0.95 for specific IEM, though external validation remains limited. Newborn screening expanded from single-disease to comprehensive panels detecting over 50 disorders. Conclusions: This comprehensive review demonstrates that IEMs have evolved from rare curiosities to systematically diagnosable conditions through technological advances. Integration of metabolomics, genomics, proteomics and artificial intelligence promises further diagnostic improvements. Standardized classification systems and evidence-based diagnostic algorithms are essential for optimal patient care. Future directions include artificial intelligence-enhanced diagnostics, expanded screening, and personalized medicine approaches.