Journal of Pediatrics vol:156 issue:5 pages:810-817
OBJECTIVES: To investigate different aspects of the introduction of array comparative genomic hybridization (aCGH) in clinical practice. STUDY DESIGN: A total 150 patients with a syndromic congenital heart defect (CHD) of unknown cause were analyzed with aCGH at 1-Mb resolution. Twenty-nine of these patients, with normal results on 1Mb aCGH, underwent re-analysis with 244-K oligo-microarray. With a logistic regression model, we assessed the predictive value of patient characteristics for causal imbalance detection. On the basis of our earlier experience and the literature, we constructed an algorithm to evaluate the causality of copy number variants. RESULTS: With 1-Mb aCGH, we detected 43 structural variants not listed as clinically neutral polymorphisms, 26 of which were considered to be causal. A systematic comparison of the clinical features of these 26 patients to the remaining 124 patients revealed dysmorphism as the only feature with a significant predictive value for reaching a diagnosis with 1-Mb aCGH. With higher resolution analysis in 29 patients, 75 variants not listed as clinically neutral polymorphisms were detected, 2 of which were considered to be causal. CONCLUSIONS: Molecular karyotyping yields an etiological diagnosis in at least 18% of patients with a syndromic CHD. Higher resolution evaluation results in an increasing number of variants of unknown significance.