Factors associated with maternal cell contamination in amniocentesis samples as evaluated by fluorescent in situ hybridization

OBJECTIVE: To determine which patient- and procedure-related factors contribute to maternal cell contamination in uncultured amniocentesis fluid. METHODS: One hundred thirty amniotic fluid (AF) samples were obtained by three operator groups: maternal-fetal medicine faculty (n=50), general obstetrician gynecologists (n=50), and obstetrics and gynecology residents supervised by maternal-fetal medicine faculty (n=30). These groups were designated "most," "intermediate," and "least experience," respectively. Study variables were recorded at the time of the procedure. Amniotic fluid cells from male fetuses underwent fluorescent in situ hybridization. Maternal cell contamination was calculated by analyzing 100 cells and determining the number of XX and XY cells. A control system was created to validate the methods used for AF processing and cell counting. RESULTS: Median maternal cell contamination was 2.0%. Maternal cell contamination did not vary with body mass index (r=-.13, P=.14), gestational age (r=.08, P=.35), or placental location (P=.55). Maternal cell contamination was significantly elevated with placental penetration (6.0% compared with 1.0%, P < .001), two passes (27.5% compared with 2.0%, P=.002), blood-tinged fluid color (14.0% compared with 2.0%, P < .001), and operator inexperience ("intermediate experience" compared with "most experience," 4.5% compared with 1.0%, P=.026). Maternal cell contamination did not differ between the "most experience" and "least experience" groups (1.0% compared with 2.0%, not significant). Concordance between detected and actual maternal cell contamination in the control system was extremely high (concordance coefficient=0.98, P=.008), confirming the validity of the techniques used. CONCLUSION: Our techniques of cell counting and maternal cell contamination calculation are accurate. Maternal cell contamination is increased with placental penetration, two passes, and operator inexperience. However, with expert supervision, inexperienced physicians can perform amniocentesis without an increase in maternal cell contamination.