Answers are grounded in this paper and cite their sources.
Prenatal diagnostic methods and screening procedures have changed rapidly in recent years. Previously, only karyotyping was performed prenatally and monitoring was done only for Down syndrome1. Since then, diagnostic possibilities have increased to include quantitative fluorescent polymerase chain reaction, fluorescence in-situ hybridization, multiplex ligation-dependent probe amplification and chromosomal microarray analysis. Simultaneously, the basis of screening criteria has evolved from maternal age1 to biochemistry2 to combined first-trimester screening3. This has resulted in diverse test results and varying indications over time. In Denmark, as in many other countries, there is a professional desire to monitor continuously prenatal diagnostic rates of all chromosomal aberrations4. There is also great public interest in monitoring Down syndrome4. We have, in Denmark, a unique ability to monitor such data using national health registries, like the Danish Cytogenetic Central register5, that feed national pre- and postnatal karyotypes into the Danish Fetal Medicine Database6. However, for those involved in registering patient data, this is a laborious process. It puts daily demands on clinicians' time, requires difficult decisions to be made on how to obtain uniform registrations, and, in spite of these decisions, there are large data cleaning chores after collection to group the data in meaningful ways. We constructed an algorithm to decrease the workload involved in the last of these steps. As part of a study on microcephaly, we obtained data on all Danish pregnancies with normal first-trimester risk assessment, no known anomalies, and that underwent a second-trimester scan including fetal biometry during the period 1 February 2008 to 1 October 2015 (n = 354 544). The data were collected prospectively on pregnancies in all 22 departments of obstetrics and gynecology in Denmark. These data were retrieved from the Danish Fetal Medicine Database, from which pre- and postnatal karyotyping results, fed in from the Danish Cytogenetic Central Register, were extracted. We then constructed an algorithm (Appendix S1) to group karyotypes into normal or abnormal, and to subdivide abnormal ones into type of aneuploidy, translocation, deletion or duplication (Table 1). Additional variables defined fetal sex, mosaicism and whether the aberration is known to have low penetrance7. After constructing this algorithm, we applied it to the entire 2008–2015 prenatal karyotype data extracted from the Danish Fetal Medicine Database. It instantly produced a grouping of results that would allow the database to monitor Down syndrome pre- and postnatally without much additional work (Table 1). Previously, a clinical geneticist would have had to review manually all abnormal karyotypes to identify trisomy 21 karyotypes for detailing in annual reports. We are, of course, aware that some of the subgroups are relatively inhomogeneous (e.g. the deletion and/or duplication subgroup). According to the International System for Human Cytogenetic Nomenclature, only karyotypes presumed abnormal on the basis of clinical findings should be reported as abnormal, whereas familial copy number variants should result in a normal karyotype. Here, variants of unknown significance, susceptibility variants and pathogenic variants are grouped together. If karyotype data are to be used for research, data cleaning should be done with more scrutiny, and often one should return to the reporting laboratory to receive full insight into the analyses performed. We hope that the algorithm can be of use in other settings after local adaptations are made, and for additional purposes other than in the Danish Fetal Medicine Database. Funding from the Novo Nordic Foundation (NNF16OC0018772) for I.V. is greatly appreciated. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
No takes yet. Share an insight, caveat, or question.
Vogel et al. (2017) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: