- Email: [email protected]
Is maternal triple screening a better predictor of down syndrome in female than male fetuses?
SPO Abstracts
Volume 176, N u m b e r 1, Part 2 Am J Obstet GynecoI
289
POTENTIAL MISINTERPRETATION OF SCREENING RISK FOR TRISOMY 2 l. L Chik, ~ K Spencer;* MPJohnson, tcJ Sokol, MPDombrowski, I Zado,; ~ MI Evans. Depts. of Ob/Gyn, Molecular Medicine & Genetics, and PathologT, WSU, Detroit, and Dept. of Clinical Biochemistly, Oldchurch Hospital, Romford, England OBJECTIVE: Maternal serum screening results are commonly expressed as risk estimates, i.e. the reciprocal of the corresponding odds. The ohjective ot this study was to determine if such presentations are open to nlisinterpretation. STUDY DESIGN: MSAFP and free 13-HCG were determined in conventional MOMs and logarithms with transfimns to compute likelihood ratios (LR) for trisomy 21 (T21) along with corresponding probabilities based upon maternal age (P) using the Glasgow ratio. 58,296 cases with 348 T21's were used. RESULTS: Statistical means (~), standard deviations (SD), and medians for T21 I.R and odds ratio show they were not normally distributed (confirmed by using W statistics)
P*LR 2
SD
291
IS THE DEGREE AND DIRECTION OF CHANGE O N REPEAT ELEVATED MATERNAL SERUM ALPHA FETOPROTEIN PREDICTIVE OF NEURAL TUBE DEFECTS OR ADVERSE PREGNANCY OUTCOME? A. Scisdone,' C E~callon/ L. Norrdl, x K. Blakemm-e. Dept. GYN/OB, Johns Hopkins School of Medicine, Baltimore, MD. OBJECTIVE: An elevated maternal serum alpha fi~toprotein (MSAFP) is associated with an increased risk of neural tube defects (NTDs) and poor pregnaocy outcome. We examined the effect of degree and direction of change beP,veen two consecutive elevated MSAFP values on the occun'ence of NTDs and adverse pregnancy outcomes. STUDY DESIGN: We reviewed the charts of patients (pts) reterred ior an elevated MSAFP from 1/86 to 1/91. Pts included had ~'o consecutive MSAFP values >2.0 M.o.M., a sonogram (u/s) that confirmed a single gestation and gestational age of 15-21 weeks. A genetic, medical, and obstetrical history was taken at the time of ret~errah ,~dl pts received a targeted u / s and amniocentesis was oticred. The pregnancies were followed until delivery. In women who had a normal u/s, the occurrence of fetal demise, growth restriction, placental abruption, preeclampsia, neonatal intensive care unit (NICU) admission, gestational age at delivery, birth weight and Apgar scores were analyzed according to the direction and degree of change of MSAFP. Fisher exact and X2 were used where appropriate. RESULTS: 192 pts entered the study. The mean first MSAFP was 3.22 M.o.M. (2.0-41.9) and the mean second MSAFP was 3.01 M.o.M. (2.0-25.1). The mean change in MSAFP was 0.21 M.o.M. (_+ 1.46). Ten pts had NTDs, There were 02 adverse pregnancy outcomes. A rise in the MSAFP values was not significantly associated with NTD or adverse pregnancy outcome. There was a significant association with fetal demise (p 0.028), NICU admission (p 0.001), and a five minute Apgar score <7 (p = 0.038) in pregnancies that had a decrease in MSAFP of -->0.75 M.o.M. No other outcome variables were significantly affected. CONCLUSION: In women with ~,o consecutive elevate MSAFP values, a decline -->0.75 M.o.M. is associated with fetal demise, NICU admission, and low Apgar scores. Degree and direction of change in MSAFP in this group of patients was not helpful in predicting fetuses with NTDs, not" preterm birth, fetal growth restriction, placental abruption, or preeclampsia,
292
IS MATERNAL TRIPLE SCREENING A BE'I'rER PREDICTOR OF DOWN SYNDROME 1N FEMALE T H A N MALE FETUSES? A. Ghidini, C.K Sports', R.E. Gri~r', C.A( Walke~~, .tic Pezzullo~. Dpt. Ob/Gyn, Georgetown University Medical Center, and Genzyme Genetics, Wasilington DC and Santa Fe, NM. OBJECTIVE: Among euploid gestations, female fetuses have been reported to have significantly lower maternal serum alpha-fetoprotein (MSMF'P) and higher human chorionic gonadotropin (hCG) levels than male fetuses. Since in the triple screening, low MSAFP and high hCG MOM independently confer greater risk of a Down syndrome fetus, we investigated the hypothesis that maternal triple screening is more etficacious at detecting female than male Down syndrome fetuses. METHODS: A database containing all karyotypes fi'om amniocenteses performed during the period 8/94 and 8/96 was accessed. Down syndrome cases due to trisomy 21 were identified. The male:fentale ratio among trisomy 21 fetuses detected after maternal serum triple screening was compared with that among trisonly 21 fetuses fi-om amniocenteses done because of advanced maternal age (AMA) that were used as controls. Statistical analysis utilized Chi-square. A p value <0.05 was considered statistically significant. RESULTS: 41 women with Down syndrome fetuses undmwent amniocentesis because ofabnormaI triple screening and 31I because of A[vL4..The male:felnale ratio among the triple screening group was not significantly diffb'rent from that of the AMA group (1:0.86 vs 1:0.75, p = 0.79). Our study had a power of 80% to detect a difference of 30% in the male:female ratio (c~ - 0.05, 13 = 0.20). CONCLUSIONS: The reported differences in MSAFP and hCG levels between male and female euploid tietuses do not appear to afli~ct the sex ratio among Down syndronte tietuses detected because of an abnormal maternal triple screening.
ODD median
2
SD - ~
ntedian
l
nls
T2I
.0014 .0595
.0097 .1801
.0003 .0064
1/4478.0 1/742,2
1/4140.1 1/85.7
1/3207 l/i57
While P*LR has a moderate correlation with T21 (r - .2557), odds have a much smaller correlation (r .0695). Sequential truncation showed the P*LR rentains little changed (r - .2542) using only the first three decimal places, but odds inore than double in correlations at 2,000 (r - -.1433); 1,000 (r = -.1776); 500 (r = -.2109). CONCLUSIONS: As the ODDS are reciprocals of small decimal numbers, there value may blow up beyond the use of a range with deteriorated statistical significance. Suggested truncations on the MOMs do not appear to guarantee reliable statistics. Our data suggest that ratios < I / 2 0 0 0 should not be reported as they can give patients a fMse sense of security not justified by actual risks.
290
IN AMN1OTIC FLUID AFP. A Drugan, XJMn*: phy/ Y Yaron, ~ SAD Ebrahim, ~ tgL K~amer,x MPJohnson, MI F.vans. Dept. of GENDER D I ~ N C E S
Ob/Gyn, Rambam Medical Center, Haifa, Israel; Depts. of Ob/Gyn, Molecular Medicine & Genetics, and PathoIogy, Wayne State Universit}, Detroit, MI. OBJECTIVE: To evaluate the effect of fetal sex on the concentration of amniotic fluid alpha-fetoprotein (AF-AFP) in singletons and twins. STUDY DESIGN: Anmiocentesis was performed tot advanced maternal age between 15 to 20 weeks of gestation. Only patients with normal karyotypes, uncomplicated gestations, and normal ultrasound examination were included. AF-AFP was measured in amniotic fluid bv radioimlnuno a~ay and resnhs, expressed as MOMs, were grouped according to fetal sex, and compared by I-test. RESULTS: A total of 603 singleton pregnancies (294 fiemales, 309 males) and 38 rain pregnancies discordant for sex met the inclusion criteria.
Singletons (n - 603) AFAFP
Mean MoM
SD
"l~ins (n = 38 pairs9 Mean MoM
I
SD
l
Females Males Probability
1.04 1.06 NS
+0.33 +0.34
1.0S 1.14 NS
+0.27 +0.39
Gender had no impact on AF-AFp in singleton pregnancies. The mean maternal age was similar in both groups (37.94 - 2.22 and 37.91 + 1.95 for male and female gestations, respectively). No effect of maternal age on AF-AFP in singletons could be documented, for either male or female tetuses (Pearson correlation matrix 0.06 and 0.03, respectively). Among twins, a slight but non-significant increase in AF-AFP was noted in males (P 0.I). CONCLUSIONS: Amniotic fluid AFP is similar in singleton gestations with male and female fetuses, suggesting that the dilierential influence of sex hormones on the activity of the AFP gene is negligible. However in twins, subtle differences can be appreciated that suggest physiologic influences of circulating androgens that are not completely accounted tor by testosterone-estradiol binding globulin that binds sex hormones.
$89
Volume 176, N u m b e r 1, Part 2 Am J Obstet GynecoI
289
POTENTIAL MISINTERPRETATION OF SCREENING RISK FOR TRISOMY 2 l. L Chik, ~ K Spencer;* MPJohnson, tcJ Sokol, MPDombrowski, I Zado,; ~ MI Evans. Depts. of Ob/Gyn, Molecular Medicine & Genetics, and PathologT, WSU, Detroit, and Dept. of Clinical Biochemistly, Oldchurch Hospital, Romford, England OBJECTIVE: Maternal serum screening results are commonly expressed as risk estimates, i.e. the reciprocal of the corresponding odds. The ohjective ot this study was to determine if such presentations are open to nlisinterpretation. STUDY DESIGN: MSAFP and free 13-HCG were determined in conventional MOMs and logarithms with transfimns to compute likelihood ratios (LR) for trisomy 21 (T21) along with corresponding probabilities based upon maternal age (P) using the Glasgow ratio. 58,296 cases with 348 T21's were used. RESULTS: Statistical means (~), standard deviations (SD), and medians for T21 I.R and odds ratio show they were not normally distributed (confirmed by using W statistics)
P*LR 2
SD
291
IS THE DEGREE AND DIRECTION OF CHANGE O N REPEAT ELEVATED MATERNAL SERUM ALPHA FETOPROTEIN PREDICTIVE OF NEURAL TUBE DEFECTS OR ADVERSE PREGNANCY OUTCOME? A. Scisdone,' C E~callon/ L. Norrdl, x K. Blakemm-e. Dept. GYN/OB, Johns Hopkins School of Medicine, Baltimore, MD. OBJECTIVE: An elevated maternal serum alpha fi~toprotein (MSAFP) is associated with an increased risk of neural tube defects (NTDs) and poor pregnaocy outcome. We examined the effect of degree and direction of change beP,veen two consecutive elevated MSAFP values on the occun'ence of NTDs and adverse pregnancy outcomes. STUDY DESIGN: We reviewed the charts of patients (pts) reterred ior an elevated MSAFP from 1/86 to 1/91. Pts included had ~'o consecutive MSAFP values >2.0 M.o.M., a sonogram (u/s) that confirmed a single gestation and gestational age of 15-21 weeks. A genetic, medical, and obstetrical history was taken at the time of ret~errah ,~dl pts received a targeted u / s and amniocentesis was oticred. The pregnancies were followed until delivery. In women who had a normal u/s, the occurrence of fetal demise, growth restriction, placental abruption, preeclampsia, neonatal intensive care unit (NICU) admission, gestational age at delivery, birth weight and Apgar scores were analyzed according to the direction and degree of change of MSAFP. Fisher exact and X2 were used where appropriate. RESULTS: 192 pts entered the study. The mean first MSAFP was 3.22 M.o.M. (2.0-41.9) and the mean second MSAFP was 3.01 M.o.M. (2.0-25.1). The mean change in MSAFP was 0.21 M.o.M. (_+ 1.46). Ten pts had NTDs, There were 02 adverse pregnancy outcomes. A rise in the MSAFP values was not significantly associated with NTD or adverse pregnancy outcome. There was a significant association with fetal demise (p 0.028), NICU admission (p 0.001), and a five minute Apgar score <7 (p = 0.038) in pregnancies that had a decrease in MSAFP of -->0.75 M.o.M. No other outcome variables were significantly affected. CONCLUSION: In women with ~,o consecutive elevate MSAFP values, a decline -->0.75 M.o.M. is associated with fetal demise, NICU admission, and low Apgar scores. Degree and direction of change in MSAFP in this group of patients was not helpful in predicting fetuses with NTDs, not" preterm birth, fetal growth restriction, placental abruption, or preeclampsia,
292
IS MATERNAL TRIPLE SCREENING A BE'I'rER PREDICTOR OF DOWN SYNDROME 1N FEMALE T H A N MALE FETUSES? A. Ghidini, C.K Sports', R.E. Gri~r', C.A( Walke~~, .tic Pezzullo~. Dpt. Ob/Gyn, Georgetown University Medical Center, and Genzyme Genetics, Wasilington DC and Santa Fe, NM. OBJECTIVE: Among euploid gestations, female fetuses have been reported to have significantly lower maternal serum alpha-fetoprotein (MSMF'P) and higher human chorionic gonadotropin (hCG) levels than male fetuses. Since in the triple screening, low MSAFP and high hCG MOM independently confer greater risk of a Down syndrome fetus, we investigated the hypothesis that maternal triple screening is more etficacious at detecting female than male Down syndrome fetuses. METHODS: A database containing all karyotypes fi'om amniocenteses performed during the period 8/94 and 8/96 was accessed. Down syndrome cases due to trisomy 21 were identified. The male:fentale ratio among trisomy 21 fetuses detected after maternal serum triple screening was compared with that among trisonly 21 fetuses fi-om amniocenteses done because of advanced maternal age (AMA) that were used as controls. Statistical analysis utilized Chi-square. A p value <0.05 was considered statistically significant. RESULTS: 41 women with Down syndrome fetuses undmwent amniocentesis because ofabnormaI triple screening and 31I because of A[vL4..The male:felnale ratio among the triple screening group was not significantly diffb'rent from that of the AMA group (1:0.86 vs 1:0.75, p = 0.79). Our study had a power of 80% to detect a difference of 30% in the male:female ratio (c~ - 0.05, 13 = 0.20). CONCLUSIONS: The reported differences in MSAFP and hCG levels between male and female euploid tietuses do not appear to afli~ct the sex ratio among Down syndronte tietuses detected because of an abnormal maternal triple screening.
ODD median
2
SD - ~
ntedian
l
nls
T2I
.0014 .0595
.0097 .1801
.0003 .0064
1/4478.0 1/742,2
1/4140.1 1/85.7
1/3207 l/i57
While P*LR has a moderate correlation with T21 (r - .2557), odds have a much smaller correlation (r .0695). Sequential truncation showed the P*LR rentains little changed (r - .2542) using only the first three decimal places, but odds inore than double in correlations at 2,000 (r - -.1433); 1,000 (r = -.1776); 500 (r = -.2109). CONCLUSIONS: As the ODDS are reciprocals of small decimal numbers, there value may blow up beyond the use of a range with deteriorated statistical significance. Suggested truncations on the MOMs do not appear to guarantee reliable statistics. Our data suggest that ratios < I / 2 0 0 0 should not be reported as they can give patients a fMse sense of security not justified by actual risks.
290
IN AMN1OTIC FLUID AFP. A Drugan, XJMn*: phy/ Y Yaron, ~ SAD Ebrahim, ~ tgL K~amer,x MPJohnson, MI F.vans. Dept. of GENDER D I ~ N C E S
Ob/Gyn, Rambam Medical Center, Haifa, Israel; Depts. of Ob/Gyn, Molecular Medicine & Genetics, and PathoIogy, Wayne State Universit}, Detroit, MI. OBJECTIVE: To evaluate the effect of fetal sex on the concentration of amniotic fluid alpha-fetoprotein (AF-AFP) in singletons and twins. STUDY DESIGN: Anmiocentesis was performed tot advanced maternal age between 15 to 20 weeks of gestation. Only patients with normal karyotypes, uncomplicated gestations, and normal ultrasound examination were included. AF-AFP was measured in amniotic fluid bv radioimlnuno a~ay and resnhs, expressed as MOMs, were grouped according to fetal sex, and compared by I-test. RESULTS: A total of 603 singleton pregnancies (294 fiemales, 309 males) and 38 rain pregnancies discordant for sex met the inclusion criteria.
Singletons (n - 603) AFAFP
Mean MoM
SD
"l~ins (n = 38 pairs9 Mean MoM
I
SD
l
Females Males Probability
1.04 1.06 NS
+0.33 +0.34
1.0S 1.14 NS
+0.27 +0.39
Gender had no impact on AF-AFp in singleton pregnancies. The mean maternal age was similar in both groups (37.94 - 2.22 and 37.91 + 1.95 for male and female gestations, respectively). No effect of maternal age on AF-AFP in singletons could be documented, for either male or female tetuses (Pearson correlation matrix 0.06 and 0.03, respectively). Among twins, a slight but non-significant increase in AF-AFP was noted in males (P 0.I). CONCLUSIONS: Amniotic fluid AFP is similar in singleton gestations with male and female fetuses, suggesting that the dilierential influence of sex hormones on the activity of the AFP gene is negligible. However in twins, subtle differences can be appreciated that suggest physiologic influences of circulating androgens that are not completely accounted tor by testosterone-estradiol binding globulin that binds sex hormones.
$89