All systematic searches for uniparental disomy (UPD) so far published and comprising clinically defined populations (Silver-Russell syndrome/primordial growth retardation (SRS/PGR) (n = 14), multiple malformations (n = 2), or rare syndromes (n = 12)) or situations at risk (confined placental mosaicism (CPM) (n = 13), spontaneous abortions (n = 6), additional marker chromosomes (n = 15), balanced non-Robertsonian translocations (n = 3), or balanced Robertsonian translocations (n = 15)) were reviewed. In many studies clinical and/or cytogenetic information on fluorescent in situ hybridization (FISH) results was very scarce. Meta-analysis concerning an adequate number of cases was possible for SRS/PGR, CPM, additional marker chromosomes, and balanced Robertsonian translocations only. As expected, the highest risk for UPD was found in cases with translocations between homologous acrocentric chromosomes (11 cases with UPD of 15 investigated) and in CPM due to a meiotic error (25 of 51 cases). In prenatal investigations or in cases with a normal phenotype, translocations between nonhomologous acrocentric chromosomes implied a risk for UPD of less than 0.5%. The risks for maternal UPD 7 in cases with SRS/PGR, for UPD 15 in cases with an additional inv dup(15) marker chromosome, and for UPD of any chromosome in cases with multiple malformation/mental retardation were approximately 5.5%, and approximately 1.3%, respectively. Searches for UPD in well-defined syndromes (Brachmann-De Lange syndrome, Sotos syndrome, Rett syndrome, Weaver syndrome, or XX true hermaphroditism) were disappointing. Not a single case was found.