Methods
This study was a secondary analysis of the Stillbirth Collaborative
Research Network (SCRN) study. Briefly, the SCRN study was a racially,
ethnically, and geographically diverse, multicenter case-control study
of stillbirth and selected live births with enrollment at the time of
delivery. Recruitment occurred in 59 hospitals in 5 geographic regions
throughout the US. Details about the participating hospitals and study
population have been described previously.11, 14 The
study was approved by the institutional review board at each clinical
site and the data coordinating center. An advisory board reviewed the
progress and safety of the study and written informed consent was
obtained from each participant.
This work was supported by grant funding from the Eunice Kennedy
Shriver National Institute of Child Health and Human Development:
U10-HD045953 Brown University, Rhode Island; U10-HD045925 Emory
University, Georgia; U10-HD045952 University of Texas Medical Branch at
Galveston, Texas; U10-HDO45955 University of Texas Health Sciences
Center at San Antonio, Texas; U10-HD045944 University of Utah Health
Sciences Center, Utah; and U01-HD045954 RTI International, RTP.
Secondary analysis of the primary research was supported in part by the
National Center for Advancing Translational Sciences of the National
Institutes of Health under Award Number 1UL01TR002538. The content is
solely the responsibility of the authors and does not necessarily
represent the official views of the National Institutes of Health.
Postmortem examinations and placental histologic examinations were
performed by perinatal pathologists who underwent centralized training
using a standardized format. Details of the placental pathologic
evaluation have been previously reported.9 The INCODE
cause of death classification tool was used across sites to best
identify cases where a fetal or placental condition significantly
contributed to the fetal death.15, 16 In the present
analysis, we included singleton stillbirth deliveries with chromosomal
microarray and postmortem examinations of the fetus and placenta
(n=387). Of note, 98.6% of stillbirth cases in the SCRN study had
placental examination completed.9.
A consensus-determined protocol was implemented to define placental
pathological lesions types as maternal vascular, fetal vascular,
maternal inflammatory, fetal inflammatory and immune/idiopathic
lesions.17, 18 The specific pathologic placental
lesions included in each category are listed in Table S1 .
Biospecimens collected as part of the SCRN protocol included placental
tissue, fetal liver, muscle, and cord blood. Sizes of placental biopsies
varied, but they were as large as 1 cm3 and were
stored at – 200C from 2 – 5 years prior to DNA
extraction. Microarray analysis was performed at a single laboratory
(Columbia University Medical Center) in 2012. Samples were analyzed
using the Affymetrix Genome Wide Human SNP Array 6.0 and the Chromosome
Analysis Suite, version 1.0.1, and the NetAffx annotation database,
version 28 with data aligned to the Human Genome release 18. CNVs of
≥500 kb in size were detected using the SNP array. Analysis of the array
data was conducted to determine aneuploidy, potential maternal-fetal
contamination, and sex discordance. Classification of CNVs was based on
the American College of Medical Genetics (ACMG) standards and guidelines
for interpretation and reporting, with modifications as described
previously.19 Due to improving resolution for
determination of pathogenicity of CNVs, the number of novel structural
variants is constantly increasing.20 Therefore, we
implemented the latest ACMG guidelines21 in
high-throughput CNV analysis to classify and update pathogenicity of
CNVs previously categorized as variants of unknown clinical significance
(VOUS) by using ClassifyCNV tool.22 Since VOUS CNVs
should not be considered benign, we classified CNVs into two groups:
abnormal CNVs (abnormal CNVs), defined as pathogenic CNVs (including
aneuploidy) or VOUS, and normal CNVs (normal CNVs), defined as no CNVs
> 500 kb or benign CNVs.23-25. As such,
the abnormal CNVs and normal CNVs groups were compared in statistical
analysis. In addition, study characteristics were described separately
for pathogenic, VOUS and normal CNVs. We discussed the implications of
the findings based on only pathogenic CNVs (excluding trisomy, monosomy,
sex-chromosome and VOUS CNVs).
To test for associations of placental pathological lesions with CNV
categorization, frequency and percentages were calculated within
category and compared using chi-square. Other categorical measures were
similarly compared by CNV category. To compare continuous measures,
ANOVA was used. Data were analyzed with the use of statistical software
programs: SAS version 9.4 (SAS Institute Inc), R and STATA version 15.0
(StataCorp), and ClassifyCNV tool.22