- Email: [email protected]
80: Impact of fetal gender on the risk of preterm birth
Oral Concurrent Session 8 Saturday, February 7, 2015
d
8:00 am e 10:00 am
d
ajog.org
Indigo AE
INFECTIOUS DISEASE
Abstracts 79 e 86 Moderators: Ed Chien, MD; Alison Stuebe, MD 79 Metagenomics of a population-based prospective cohort reveals no discernable impact of mode of delivery (MOD) on the infant microbiome by 6 weeks of age
Figure Infant microbiome clusters predominately by body site, not MOD by 6 weeks
Derrick Chu1, Jun Ma1, Amanda Prince1, Kathleen Antony1, Michelle Moller1, Brigid Boggan1, Kjersti Aagaard1 1
Baylor College of Medicine, Houston, TX
OBJECTIVE: The infant microbiome is relatively sparse at birth, and
populated early in life. The reported relative impact of MOD on the neonatal microbiome has garnered attention due to Cesarean (CD) prevalence and its association with later in life disease. Although previous small studies suggested that CD adversely impacted microbiome diversity, they were limited by potential confounding and lack of longitudinal sampling. We thus sought to interrogate whether MOD impacts the infant microbiome over time in a robust population-based cohort. STUDY DESIGN: An at-large representative cohort (n¼277 gravidae) was prospectively enrolled, and a subset (n¼81) consented to longitudinal sampling (3rd trimester, delivery & 4-6 wks postpartum). Non-contaminated samples were uniformly collected at 5 infant sites (meconium/stool, oral supragingival plaque, respiratory anterior nares, and skin auricular & antecubital fossa) and 7 maternal sites (aforementioned plus vaginal introitus and posterior fornix). DNA was extracted (MolBio), NextGen sequenced, and subjected to robust comprehensive 16S and WGS-based metagenomic analysis. RESULTS: High quality sequence data (>3Tb) was obtained, with complete longitudinal maternal and infant sample sets in a nested cohort (22 CD & 52 VB). At delivery, the infant microbiome was significantly different between CD and VB infants along the first principal component (PC1) axis for all body sites (panel A; MannWhitney *p<0.05,**p<0.01). However, by 6 weeks the infant microbiome clusters by body site (panel B; PERMANOVA p¼0.001) but not MOD (PERMANOVA p¼0.073 & PC1 Mann-Whitney, all p>0.20). CONCLUSION: Although MOD marginally influenced the infants microbiome at delivery, there was no lasting discernable impact by 6 weeks of age. Based on our prior findings, we speculate that other factors (i.e., the placental microbiome, gestational age & breastfeeding) may bear greater effect on establishment and early development of the infant microbiome.
Principal coordinates analysis (PCoA) of infants either at (A) Delivery or (B) 4-6 weeks of age. The value of the first prinicipal component (PC1) for each sample is shown, separated by body site and by MOD. There is a significant difference between CD and VB along PC1 for all body sites (*p<0.05,**p<0.01) at delivery, but not at 6 weeks of age (all p>0.20).
80 Impact of fetal gender on the risk of preterm birth
Myrthe Peelen1, Brenda Kazemier1, Anita Ravelli1, Christianne de Groot2, Petra Hajenius1, Ben Mol3, Marjolein Kok1 1 Academic Medical Center, Amsterdam, Netherlands, 2VU Medical Center, Department of Obstetrics and Gynaecology, Amsterdam, Netherlands, 3 University of Adelaide, The Robinson Institute, School of Paediatrics and Reproductive Health, Adelaide, SA, Australia
OBJECTIVE: To study the influence of fetal gender on preterm birth
and neonatal outcome in Caucasian women. STUDY DESIGN: We performed a national cohort study using data
collected in the Netherlands Perinatal Registry. The study population comprised 1,947,266 singleton births (25+0-42+6 weeks) from Caucasian women between 1999 and 2010 with a fetus alive at the onset of labour. Cases with congenital anomalies or unknown fetal
S56 American Journal of Obstetrics & Gynecology Supplement to JANUARY 2015
ajog.org
Infectious Disease
gender were excluded. The relative risk ratios for gender per week of gestation were assessed as well as gender related risk on adverse neonatal outcomes. These outcomes were perinatal mortality and a composite of neonatal morbidity (defined as neonatal intensive care admission, sepsis, meconium aspiration, necrotizing enterocolitis, respiratory distress syndrome or intraventricular haemorrhage). Onset of labor was categorized in spontaneous onset with intact membranes, prelabor rupture of membranes (PROM) and induction of labor/elective caesarean section. We used a moving average technique covering 3 weeks per measurement to correct for possible fluctuations due to the small number of events. RESULTS: Male fetuses were at increased risk of spontaneous preterm birth with intact membranes compared to a female fetus with a peak between 28 and 31 weeks (RR 1.5 for male to female; 95% CI 1.31.7). In addition, male fetuses were at increased risk of PPROM between 26 and 37 weeks (RR 1.2; 95 CI 1.16-1.23). No gender effect was seen for medically indicated preterm birth. There were no significant differences between male and female fetuses born at comparable gestational ages regarding neonatal mortality. However, males were at significantly increased risk of composite neonatal morbidity compared to females from 29 weeks onwards with a peak at 37-38 weeks (RR 1.4; 95% CI 1.2-1.5). CONCLUSION: Male fetal gender is an important risk factor for spontaneous preterm birth, both for intact membranes and for PPROM. In addition, males are at increased risk of neonatal morbidity mainly in the term period.
Oral Concurrent Session 8
but increased the MBR 22- to 26-fold (20.9-32.6% vs 0.8-1.5%). The cumulative LBR over two cycles with 1 ET was as good or better than 2 ET in a single cycle (for women age 20: 42.2% vs 38.4%; for women age 40, 31.6% vs 31.5%), while greatly reducing the MBR (bolded values). CONCLUSION: The cumulative LBR is as good or better with 1 ET over two cycles than 2 ET in a single cycle, and the probability of a multiple birth is greatly reduced.
82 Detection of microbial invasion of the amniotic cavity by analysis of cervicovaginal proteins in preterm labor with intact membranes
C Andrew Combs1, Jodi Lapidus2, Michael Gravett3, Thomas Garite4, Jerome Lapointe5, for the ProteoGenix/Obstetrix Collaborative Research Network6
1 Obstetrix Medical Group, Campbell, CA, 2Oregon Health and Science University, Portland, OR, 3University of Washington, Seattle, WA, 4Obstetrix and University of California, Irvine, CA, 5Hologic, Inc, Sunnyvale, CA, 6 Obstetrix Center for Research, Education, and Quality, ProteoGenix, Costa Mesa, CA, Sunrise, FL
OBJECTIVE: Microbial invasion of the amniotic cavity (MIAC) is
81 The effect of transferring fewer embryos over more cycles: a prediction model for live birth and multiple births with in vitro fertilization
Barbara Luke1, Morton Brown3, Ethan Wantman4, Judy Stern2, Valerie Baker5, Eric Widra6, Charles Coddington7, William Gibbons8, G. Ball9
1 Michigan State University, East Lansing, MI, 2Geisel School of Medicine at Dartmouth, Lebanon, NH, 3University of Michigan, Ann Arbor, MI, 4 Redshift Technologies, Inc., New York, NY, 5Stanford University, Palo Alto, CA, 6Shady Grove Fertility Center, Washington, DC, 7Mayo Clinic, Rochester, MN, 8Baylor College of Medicine, Houston, TX, 9Seattle Reproductive Medicine, Seattle, WA
OBJECTIVE: To develop a model predictive of live birth rates (LBR,%) and multiple birth rates (MBR,%) for women considering in vitro fertilization (IVF), based on national linked cycles from the Society for Assisted Reproductive Technology Clinic Online Reporting System for 2004-12 STUDY DESIGN: Longitudinal cohort study of women without prior IVF treatment undergoing their first two fresh autologous cycles. LBRs and MBRs were modeled by woman’s age, body mass index (BMI), gravidity, prior full-term births, and infertility diagnoses, by number of embryos transferred (ET), and cycle number, using backward-stepping logistic regression. RESULTS: The study population included 348,394 women with an initial cycle, of whom 158,983 did not become pregnant and had a second cycle. The LBRs and MBRs examples shown below are for women with a single infertility diagnosis, BMI of 20, gravidity of zero, ages 20 and 40, by cycle number (1 or 2) and ET (1 or 2). In the first cycle, 2 ET vs 1 ET improved the LBR by 8-11 percentage points,
common in early preterm labor (PTL) and is associated with maternal and neonatal infectious morbidity. MIAC is often occult and is reliably detected only with amniocentesis. We sought to develop a noninvasive test to predict MIAC based on protein biomarkers in cervicovaginal fluid (CVF) in a cohort with PTL and to validate the test in an independent cohort. STUDY DESIGN: Prospective study of women with PTL who had amniocentesis to screen for MIAC. MIAC was defined as a positive culture and/or 16S-rDNA by PCR. Nine candidate CVF proteins were analyzed by ELISA. Logistic regression was used to identify combinations of up to 3 CVF proteins that could accurately classify the Phase 1 cohort (N¼108) into those with or without MIAC. The best models were selected based on ROC area-under-curve (AUC) in Phase 1. Model performance was then tested in a second cohort (Phase 2, N¼306). RESULTS: MIAC was present in 15% of cases in Phase 1, 9% in Phase 2. A 3-marker CVF model using IL6, Gro-a, and IGF-BP1 had AUC 0.87 in Phase 1 and AUC 0.78 in Phase 2. Two-marker CVF models using IL6+Gro-a or AFP+Gro-a performed similarly in Phase 2 (AUC 0.78, 0.75 respectively), but were not superior to CVF-IL6 alone (AUC 0.80, Figure). A cut-off value of CVF-IL6 >463 pg/ml (which had 81% sensitivity in Phase 1) predicted MIAC in Phase 2 with sensitivity 79%, specificity 78%, PPV 38%, NPV 97%, +LR 3.6, -LR 0.27. CVF-IL6 was also a strong predictor of severely elevated amniotic fluid IL6 (11.3 ng/mL) in phase 2 (AUC 0.83). CONCLUSION: Cervicovaginal fluid IL-6, alone or in combination with Gro-a, is strongly associated with MIAC. If developed into a bedside test or rapid lab assay, a test based on cervicovaginal IL-6 might be useful in selecting patients in whom the probability of MIAC is high enough to warrant amniocentesis or transfer to a higher level of care. Such a test might also guide selection of potential subjects for treatment trials.
Supplement to JANUARY 2015 American Journal of Obstetrics & Gynecology
S57
d
8:00 am e 10:00 am
d
ajog.org
Indigo AE
INFECTIOUS DISEASE
Abstracts 79 e 86 Moderators: Ed Chien, MD; Alison Stuebe, MD 79 Metagenomics of a population-based prospective cohort reveals no discernable impact of mode of delivery (MOD) on the infant microbiome by 6 weeks of age
Figure Infant microbiome clusters predominately by body site, not MOD by 6 weeks
Derrick Chu1, Jun Ma1, Amanda Prince1, Kathleen Antony1, Michelle Moller1, Brigid Boggan1, Kjersti Aagaard1 1
Baylor College of Medicine, Houston, TX
OBJECTIVE: The infant microbiome is relatively sparse at birth, and
populated early in life. The reported relative impact of MOD on the neonatal microbiome has garnered attention due to Cesarean (CD) prevalence and its association with later in life disease. Although previous small studies suggested that CD adversely impacted microbiome diversity, they were limited by potential confounding and lack of longitudinal sampling. We thus sought to interrogate whether MOD impacts the infant microbiome over time in a robust population-based cohort. STUDY DESIGN: An at-large representative cohort (n¼277 gravidae) was prospectively enrolled, and a subset (n¼81) consented to longitudinal sampling (3rd trimester, delivery & 4-6 wks postpartum). Non-contaminated samples were uniformly collected at 5 infant sites (meconium/stool, oral supragingival plaque, respiratory anterior nares, and skin auricular & antecubital fossa) and 7 maternal sites (aforementioned plus vaginal introitus and posterior fornix). DNA was extracted (MolBio), NextGen sequenced, and subjected to robust comprehensive 16S and WGS-based metagenomic analysis. RESULTS: High quality sequence data (>3Tb) was obtained, with complete longitudinal maternal and infant sample sets in a nested cohort (22 CD & 52 VB). At delivery, the infant microbiome was significantly different between CD and VB infants along the first principal component (PC1) axis for all body sites (panel A; MannWhitney *p<0.05,**p<0.01). However, by 6 weeks the infant microbiome clusters by body site (panel B; PERMANOVA p¼0.001) but not MOD (PERMANOVA p¼0.073 & PC1 Mann-Whitney, all p>0.20). CONCLUSION: Although MOD marginally influenced the infants microbiome at delivery, there was no lasting discernable impact by 6 weeks of age. Based on our prior findings, we speculate that other factors (i.e., the placental microbiome, gestational age & breastfeeding) may bear greater effect on establishment and early development of the infant microbiome.
Principal coordinates analysis (PCoA) of infants either at (A) Delivery or (B) 4-6 weeks of age. The value of the first prinicipal component (PC1) for each sample is shown, separated by body site and by MOD. There is a significant difference between CD and VB along PC1 for all body sites (*p<0.05,**p<0.01) at delivery, but not at 6 weeks of age (all p>0.20).
80 Impact of fetal gender on the risk of preterm birth
Myrthe Peelen1, Brenda Kazemier1, Anita Ravelli1, Christianne de Groot2, Petra Hajenius1, Ben Mol3, Marjolein Kok1 1 Academic Medical Center, Amsterdam, Netherlands, 2VU Medical Center, Department of Obstetrics and Gynaecology, Amsterdam, Netherlands, 3 University of Adelaide, The Robinson Institute, School of Paediatrics and Reproductive Health, Adelaide, SA, Australia
OBJECTIVE: To study the influence of fetal gender on preterm birth
and neonatal outcome in Caucasian women. STUDY DESIGN: We performed a national cohort study using data
collected in the Netherlands Perinatal Registry. The study population comprised 1,947,266 singleton births (25+0-42+6 weeks) from Caucasian women between 1999 and 2010 with a fetus alive at the onset of labour. Cases with congenital anomalies or unknown fetal
S56 American Journal of Obstetrics & Gynecology Supplement to JANUARY 2015
ajog.org
Infectious Disease
gender were excluded. The relative risk ratios for gender per week of gestation were assessed as well as gender related risk on adverse neonatal outcomes. These outcomes were perinatal mortality and a composite of neonatal morbidity (defined as neonatal intensive care admission, sepsis, meconium aspiration, necrotizing enterocolitis, respiratory distress syndrome or intraventricular haemorrhage). Onset of labor was categorized in spontaneous onset with intact membranes, prelabor rupture of membranes (PROM) and induction of labor/elective caesarean section. We used a moving average technique covering 3 weeks per measurement to correct for possible fluctuations due to the small number of events. RESULTS: Male fetuses were at increased risk of spontaneous preterm birth with intact membranes compared to a female fetus with a peak between 28 and 31 weeks (RR 1.5 for male to female; 95% CI 1.31.7). In addition, male fetuses were at increased risk of PPROM between 26 and 37 weeks (RR 1.2; 95 CI 1.16-1.23). No gender effect was seen for medically indicated preterm birth. There were no significant differences between male and female fetuses born at comparable gestational ages regarding neonatal mortality. However, males were at significantly increased risk of composite neonatal morbidity compared to females from 29 weeks onwards with a peak at 37-38 weeks (RR 1.4; 95% CI 1.2-1.5). CONCLUSION: Male fetal gender is an important risk factor for spontaneous preterm birth, both for intact membranes and for PPROM. In addition, males are at increased risk of neonatal morbidity mainly in the term period.
Oral Concurrent Session 8
but increased the MBR 22- to 26-fold (20.9-32.6% vs 0.8-1.5%). The cumulative LBR over two cycles with 1 ET was as good or better than 2 ET in a single cycle (for women age 20: 42.2% vs 38.4%; for women age 40, 31.6% vs 31.5%), while greatly reducing the MBR (bolded values). CONCLUSION: The cumulative LBR is as good or better with 1 ET over two cycles than 2 ET in a single cycle, and the probability of a multiple birth is greatly reduced.
82 Detection of microbial invasion of the amniotic cavity by analysis of cervicovaginal proteins in preterm labor with intact membranes
C Andrew Combs1, Jodi Lapidus2, Michael Gravett3, Thomas Garite4, Jerome Lapointe5, for the ProteoGenix/Obstetrix Collaborative Research Network6
1 Obstetrix Medical Group, Campbell, CA, 2Oregon Health and Science University, Portland, OR, 3University of Washington, Seattle, WA, 4Obstetrix and University of California, Irvine, CA, 5Hologic, Inc, Sunnyvale, CA, 6 Obstetrix Center for Research, Education, and Quality, ProteoGenix, Costa Mesa, CA, Sunrise, FL
OBJECTIVE: Microbial invasion of the amniotic cavity (MIAC) is
81 The effect of transferring fewer embryos over more cycles: a prediction model for live birth and multiple births with in vitro fertilization
Barbara Luke1, Morton Brown3, Ethan Wantman4, Judy Stern2, Valerie Baker5, Eric Widra6, Charles Coddington7, William Gibbons8, G. Ball9
1 Michigan State University, East Lansing, MI, 2Geisel School of Medicine at Dartmouth, Lebanon, NH, 3University of Michigan, Ann Arbor, MI, 4 Redshift Technologies, Inc., New York, NY, 5Stanford University, Palo Alto, CA, 6Shady Grove Fertility Center, Washington, DC, 7Mayo Clinic, Rochester, MN, 8Baylor College of Medicine, Houston, TX, 9Seattle Reproductive Medicine, Seattle, WA
OBJECTIVE: To develop a model predictive of live birth rates (LBR,%) and multiple birth rates (MBR,%) for women considering in vitro fertilization (IVF), based on national linked cycles from the Society for Assisted Reproductive Technology Clinic Online Reporting System for 2004-12 STUDY DESIGN: Longitudinal cohort study of women without prior IVF treatment undergoing their first two fresh autologous cycles. LBRs and MBRs were modeled by woman’s age, body mass index (BMI), gravidity, prior full-term births, and infertility diagnoses, by number of embryos transferred (ET), and cycle number, using backward-stepping logistic regression. RESULTS: The study population included 348,394 women with an initial cycle, of whom 158,983 did not become pregnant and had a second cycle. The LBRs and MBRs examples shown below are for women with a single infertility diagnosis, BMI of 20, gravidity of zero, ages 20 and 40, by cycle number (1 or 2) and ET (1 or 2). In the first cycle, 2 ET vs 1 ET improved the LBR by 8-11 percentage points,
common in early preterm labor (PTL) and is associated with maternal and neonatal infectious morbidity. MIAC is often occult and is reliably detected only with amniocentesis. We sought to develop a noninvasive test to predict MIAC based on protein biomarkers in cervicovaginal fluid (CVF) in a cohort with PTL and to validate the test in an independent cohort. STUDY DESIGN: Prospective study of women with PTL who had amniocentesis to screen for MIAC. MIAC was defined as a positive culture and/or 16S-rDNA by PCR. Nine candidate CVF proteins were analyzed by ELISA. Logistic regression was used to identify combinations of up to 3 CVF proteins that could accurately classify the Phase 1 cohort (N¼108) into those with or without MIAC. The best models were selected based on ROC area-under-curve (AUC) in Phase 1. Model performance was then tested in a second cohort (Phase 2, N¼306). RESULTS: MIAC was present in 15% of cases in Phase 1, 9% in Phase 2. A 3-marker CVF model using IL6, Gro-a, and IGF-BP1 had AUC 0.87 in Phase 1 and AUC 0.78 in Phase 2. Two-marker CVF models using IL6+Gro-a or AFP+Gro-a performed similarly in Phase 2 (AUC 0.78, 0.75 respectively), but were not superior to CVF-IL6 alone (AUC 0.80, Figure). A cut-off value of CVF-IL6 >463 pg/ml (which had 81% sensitivity in Phase 1) predicted MIAC in Phase 2 with sensitivity 79%, specificity 78%, PPV 38%, NPV 97%, +LR 3.6, -LR 0.27. CVF-IL6 was also a strong predictor of severely elevated amniotic fluid IL6 (11.3 ng/mL) in phase 2 (AUC 0.83). CONCLUSION: Cervicovaginal fluid IL-6, alone or in combination with Gro-a, is strongly associated with MIAC. If developed into a bedside test or rapid lab assay, a test based on cervicovaginal IL-6 might be useful in selecting patients in whom the probability of MIAC is high enough to warrant amniocentesis or transfer to a higher level of care. Such a test might also guide selection of potential subjects for treatment trials.
Supplement to JANUARY 2015 American Journal of Obstetrics & Gynecology
S57