- Email: [email protected]
77: Electronic fetal heart rate monitoring and infant mortality: a population-based study in the United States
www.AJOG.org Infectious Disease/Intrapartum Fetal Assessment hours of membrane rupture. Logistic regression demonstrated only VL ⬎1000 copies/mL was an independent risk factor for perinatal HIV transmission (p⬍0.001). CONCLUSIONS: Duration of membrane rupture ⱖ4 hours is not a risk factor for perinatal transmission of HIV in women with VL ⬍1000 copies/mL on combination ART and these patients do not necessarily need to be delivered by Cesarean section.
75 Influenza-like illness in hospitalized pregnant and immediately postpartum women during the 2009-2010 H1N1 influenza pandemic Michael Varner1 1 For the Eunice Kennedy Shriver National Institute of Child Health and Human Development MaternalFetal Medicine Units Network, Bethesda, MD
OBJECTIVE: To describe characteristics and outcomes of pregnant and immediately postpartum women with influenza-like illness (ILI) during the 2009-2010 influenza pandemic in the United States. STUDY DESIGN: This is a prospective study of pregnant and postpartum (within 2 weeks of delivery) women with an initial ILI hospitalization or observation ⬎ 12 hours. ILI was defined as clinical suspicion of influenza and either meeting the CDC definition of ILI (fever ⬎ 100.0 ° F and cough/sore throat) or positive influenza test. The study was conducted at 28 hospitals. RESULTS: From October 1, 2009 through May 31, 2010, 357 women met the case definition. Thirty-five were admitted to the ICU and 4 died. Sixty-one percent were in the third trimester and only 10 were postpartum. Over half (55%) were in October and an additional 25% were in November, with rapidly decreasing numbers thereafter. In October there were 14.7 non-ICU hospital admissions/observations per 1000 deliveries and 1.4 ICU admissions per 1000 deliveries. Antivirals were administered to 8% of cases prior to hospitalization and to 88% during hospitalization. The Table lists the characteristics significantly associated with the need for ICU admission; trimester of illness, admission BMI, and asthma prevalence were not significant. Delivery outcomes are currently available for 86% of cases. CONCLUSIONS: Pregnant women with ILI were most frequently hospitalized in the fall of 2009, prior to the usual influenza season. Of those hospitalized, 1.1% died. Admission to the ICU was more common if women smoked, had pneumonia in the preceding two weeks or had chronic hypertension, and was less common in women who received antivirals within two days of symptom onset. Table: Comparison of cases admitted to ICU with those not admitted to ICU ICU (%) Cigarette smoker
29%
Non-ICU (%) 13%
Odds Ratio (95% CI)* 3.2 (1.2-8.2)
P-value* 0.02
..........................................................................................................................................................................................
Pneumonia in preceding two 11% ⬍1% 39.6 (**) 0.004 weeks .......................................................................................................................................................................................... Chronic hypertension 17% 3% 7.7 (2.0-28.3) 0.003 .......................................................................................................................................................................................... Antiviral treatment within 29% 55% 0.3 (0.1-0.8) 0.01 two days of symptom onset .......................................................................................................................................................................................... * Based on exact logistic regression including all variables in the table ** Small numbers prohibit estimation of the confidence interval
76 Intra-amniotic inflammation (IAI) in response to infection with single species versus mixed bacterial flora in women with preterm birth (PTB) Antonette Dulay1, Irina Buhimschi1, Guomao Zhao1, Unzila Ali1, Sonya Abdel-Razeq1, Sarah Lee1, Christina Han1, Lydia Shook1, Catalin Buhimschi1 1
Yale University, New Haven, CT
OBJECTIVE: Toll-like receptors (TLR) mitigate innate immune re-
Oral Concurrent Session 7
negative (GN, i.e. E. Coli) bacterial ligands. We propose that the crosstalk between TLRs generates differential IAI responses. Our aim was to evaluate the intensity of IAI elicited by single vs. mixed spp. in vivo and in vitro. STUDY DESIGN: We studied amniotic fluid (AF) from 134 women (GA: 26⫾3 wks) with PTB. AF cultures tested positive for: 1) single GN (n⫽30); 2) single GP (n⫽31); 3) Mycoplasma or Ureaplasma (n⫽34); 4) mixed TLR2 (n⫽16) or 5) mixed TLR2 & TLR4-engaging microbes. (n⫽23). In vivo, IAI level was assessed by IL6 ELISA. An amniochorion explant system (term CS, n⫽4) was used to test the effect of prototype ligands in vitro. We generated dose response curves at 6 and 18h after challenge with ligands for TLR4 [LPS], TLR2 [Lipoteichoic acid (LTA), Pam3Cys (PAM), Mycoplasma FSL] alone or combined [LPS⫹PAM, LPS⫹FSL]. TNF␣, IL6, IL8 were measured in explant media and normalized to total protein. Tissue viability was confirmed by LDH release. RESULTS: In vivo, IAI with single GN had highest AF IL6 independent of GA (p⬍0.001). There was no difference in IL6 among single TLR2engaging spp. In contrast, mixed GP induced higher IL6 vs. single TLR2-engaging spp. (p⫽0.016). IAI with mixed GN & GP had higher IL6 vs. single and mixed GP (p⬍0.05). In vitro, LPS induced higher responses in all cytokines at all times and doses while FSL responses were lowest (p⬍0.001). Of all TLR2 ligands, LTA had the highest response (p⬍0.05). LPS⫹FSL produced a 3-fold increase in all cytokines over FSL alone (p⫽0.03). LPS⫹PAM induced a 20-fold increase over PAM alone (p⫽0.02) and a transient 4-fold increase over LPS alone at 6h (p⬍0.05). CONCLUSIONS: Both in vivo and in vitro, single TLR4-engaging bacteria generate highest inflammatory responses. Mixed infections lead to different IAI levels vs. single species alone.
77 Electronic fetal heart rate monitoring and infant mortality: a population-based study in the United States Han-Yang Chen1, Suneet Chauhan2, Cande Ananth3, Anthony Vintzileos4, Alfred Abuhamad2 1
Center for Urban Population Health, Milwaukee, WI, 2Eastern Virginia Medical School, Norfolk, VA, 3UMDNJ-Robert Wood Johnson Medical School, New Brunswick, NJ, 4Stony BrookWinthrop University Hospitals, Long Island, NY
OBJECTIVE: Evidence form randomized trials demonstrate an increase in intrapartum (IP) interventions due to electronic fetal heart rate monitoring (EFM) with virtually no improvement in perinatal mortality. Yet observational (“reality-based evidence,” Vintzileos Obstet Gynecol 2009) indicates a decrease in IP mortality. We examined the association between EFM during labor and corrected neonatal and infant mortality (CIM) in the US. STUDY DESIGN: We utilized the 2004 National Birth Cohortlinked birth and infant death data file restricted to singleton births with 1989 revision of the standard certificate of live birth, used by 41 states and DC. We excluded multiple gestations, anomalous newborns, foreign residents, gestational age (GA) ⬍24 weeks or ⱖ45 weeks, implausible birthweights, or if EFM use was unknown. Multivariable log-binomial regression models were fitted to estimate risk ratio (RR) and 99% confidence interval (CI) to evaluate the association between EFM and CIM, while adjusting for potential confounders: age, race, marital Status; education; smoking and, infant’s gender. RESULTS: There were 4,112,053 live births in 2004 and 47% (n⫽1,945,789) met the inclusion criteria; among them 88% had EFM. The risks of CIM were 3.0 per 1,000 births for those with and 3.8 per 1,000 births without EFM. The decrease in CIM varies by GA (Table I). CONCLUSIONS: In the U.S., the real life utilization of EFM was associated with decreased CIM. The magnitude of this decrease varied slightly by GA and it was exclusively driven by lower number of deaths during the early neonatal period.
sponses to infection. Gram positive (GP), Mycoplasma and Ureaplasma spp. primarily engage TLR2, while TLR4 recognizes Gram Supplement to JANUARY 2011 American Journal of Obstetrics & Gynecology
S43
Oral Concurrent Session 7
Infectious Disease/Intrapartum Fetal Assessment
Table I Association between EFM and corrected infant mortality: Estimated risk ratio (99% CI) Early (0-6 days) All
Late (7-27 days)
0.54 (0.46, 0.64) 0.90 (0.70, 1.15)
Postneonatal (28-364 days)
Infant (0-364 days)
0.94 (0.82, 1.08) 0.78 (0.71, 0.86)
...........................................................................................................................................................................................................
24-27 0.64 (0.53, 0.78) 0.93 (0.64, 1.34) 1.04 (0.71, 1.55) 0.79 (0.68, 0.91) ........................................................................................................................................................................................................... 28-31 0.52 (0.35, 0.78) 1.25 (0.59, 2.64) 1.05 (0.62, 1.77) 0.78 (0.59, 1.04) ........................................................................................................................................................................................................... 32-33 0.41 (0.19, 0.90) 0.99 (0.25, 3.88) 0.94 (0.45, 1.96) 0.72 (0.44, 1.17) ........................................................................................................................................................................................................... 34-36 0.51 (0.25, 1.03) 1.02 (0.45, 2.33) 1.12 (0.73, 1.73) 0.96 (0.69, 1.33) ........................................................................................................................................................................................................... 37-40 0.67 (0.43, 1.04) 0.74 (0.48, 1.15) 0.88 (0.74, 1.06) 0.84 (0.72, 0.98) ........................................................................................................................................................................................................... ⱖ41 0.65 (0.27, 1.58) 2.23 (0.48, 10.29) 1.05 (0.66, 1.66) 1.04 (0.70, 1.54) ........................................................................................................................................................................................................... Bold if significantly different
78 Evaluation of the inter-observer and intra-observer reliability of the NICHD 3-tier FHR interpretation system Sean Blackwell1, William A. Grobman2, Leah Antoniewicz1, Maria Hutchinson1, Cynthia Gyamfi3 1 University of Texas Health Science Center at Houston, Houston, TX, 2Northwestern University, Feinberg School of Medicine, Chicago, IL, 3Columbia University, New York, NY
OBJECTIVE: In 2008, a NICHD-sponsored workshop on Electronic Fetal Monitoring (EFM) recommended adoption of a 3-Tier system for the categorization of Fetal Heart Rate (FHR) patterns. Our purpose was to test the reliability of this new classification system. STUDY DESIGN: Three MFM specialists who participated in the NICHD workshop independently reviewed FHR tracings without clinical data. All FHR tracings were from women with term, singleton pregnancies who had known umbilical artery pH results. In order to maximize the potential for a variation in FHR patterns, we included
S44
www.AJOG.org
FHR tracings from women with umbilical pH ⬍ 7.00 (n⫽6), pH 7.00-7.10 (n⫽6), and pH ⬎ 7.10 (n⫽6). FHR tracings were divided into 15-20 minute epochs and random samples were selected. Reviewers were asked to classify the FHR tracing by NICHD category (Category I, II, or III) as well as describe the individual FHR parameters. Approximately 25% of FHR tracing were duplicated and reviewed in order to evaluate intra-observer agreement. Pre-defined criteria for agreement were used: kappa 0.0 - 0.20 (poor), 0.21-0.40 (fair), 0.410.6 (moderate), 0.61-0.8 (substantial), and 0.81-1.0 (almost perfect). RESULTS: Each reviewer evaluated 154 individual FHR segments. The inter-observer reliability for the 3 examiners by each FHR category and overall is described in the table. The intra-observer agreement ranged from substantial to perfect for the 3 examiners (kappa 0.741.0). There were no cases where a FHR tracing was classified as Category I by one examiner but Category III by another. Moderate interobserver agreement was present for individual FHR parameters: presence of FHR accelerations (kappa 0.58), early decelerations (kappa 0.49), variable decelerations (kappa 0.48), late decelerations (kappa 0.54), and FHR variability (overall kappa 0.57; absent 0.16, minimal 0.53, and moderate 0.69). CONCLUSIONS: Inter-observer agreement of 3-Tier FHR classification system was moderate for Categories I and II. Agreement for Category III tracings was poor mainly due to lack of agreement regarding absent variability. FHR category
Kappa
Level of Agreement
Category I
0.53
Moderate
Category II
0.46
Moderate
.......................................................................................................................................................................................... ..........................................................................................................................................................................................
Category III ⫺0.01 Poor .......................................................................................................................................................................................... Overall 0.47 Moderate ..........................................................................................................................................................................................
American Journal of Obstetrics & Gynecology Supplement to JANUARY 2011
75 Influenza-like illness in hospitalized pregnant and immediately postpartum women during the 2009-2010 H1N1 influenza pandemic Michael Varner1 1 For the Eunice Kennedy Shriver National Institute of Child Health and Human Development MaternalFetal Medicine Units Network, Bethesda, MD
OBJECTIVE: To describe characteristics and outcomes of pregnant and immediately postpartum women with influenza-like illness (ILI) during the 2009-2010 influenza pandemic in the United States. STUDY DESIGN: This is a prospective study of pregnant and postpartum (within 2 weeks of delivery) women with an initial ILI hospitalization or observation ⬎ 12 hours. ILI was defined as clinical suspicion of influenza and either meeting the CDC definition of ILI (fever ⬎ 100.0 ° F and cough/sore throat) or positive influenza test. The study was conducted at 28 hospitals. RESULTS: From October 1, 2009 through May 31, 2010, 357 women met the case definition. Thirty-five were admitted to the ICU and 4 died. Sixty-one percent were in the third trimester and only 10 were postpartum. Over half (55%) were in October and an additional 25% were in November, with rapidly decreasing numbers thereafter. In October there were 14.7 non-ICU hospital admissions/observations per 1000 deliveries and 1.4 ICU admissions per 1000 deliveries. Antivirals were administered to 8% of cases prior to hospitalization and to 88% during hospitalization. The Table lists the characteristics significantly associated with the need for ICU admission; trimester of illness, admission BMI, and asthma prevalence were not significant. Delivery outcomes are currently available for 86% of cases. CONCLUSIONS: Pregnant women with ILI were most frequently hospitalized in the fall of 2009, prior to the usual influenza season. Of those hospitalized, 1.1% died. Admission to the ICU was more common if women smoked, had pneumonia in the preceding two weeks or had chronic hypertension, and was less common in women who received antivirals within two days of symptom onset. Table: Comparison of cases admitted to ICU with those not admitted to ICU ICU (%) Cigarette smoker
29%
Non-ICU (%) 13%
Odds Ratio (95% CI)* 3.2 (1.2-8.2)
P-value* 0.02
..........................................................................................................................................................................................
Pneumonia in preceding two 11% ⬍1% 39.6 (**) 0.004 weeks .......................................................................................................................................................................................... Chronic hypertension 17% 3% 7.7 (2.0-28.3) 0.003 .......................................................................................................................................................................................... Antiviral treatment within 29% 55% 0.3 (0.1-0.8) 0.01 two days of symptom onset .......................................................................................................................................................................................... * Based on exact logistic regression including all variables in the table ** Small numbers prohibit estimation of the confidence interval
76 Intra-amniotic inflammation (IAI) in response to infection with single species versus mixed bacterial flora in women with preterm birth (PTB) Antonette Dulay1, Irina Buhimschi1, Guomao Zhao1, Unzila Ali1, Sonya Abdel-Razeq1, Sarah Lee1, Christina Han1, Lydia Shook1, Catalin Buhimschi1 1
Yale University, New Haven, CT
OBJECTIVE: Toll-like receptors (TLR) mitigate innate immune re-
Oral Concurrent Session 7
negative (GN, i.e. E. Coli) bacterial ligands. We propose that the crosstalk between TLRs generates differential IAI responses. Our aim was to evaluate the intensity of IAI elicited by single vs. mixed spp. in vivo and in vitro. STUDY DESIGN: We studied amniotic fluid (AF) from 134 women (GA: 26⫾3 wks) with PTB. AF cultures tested positive for: 1) single GN (n⫽30); 2) single GP (n⫽31); 3) Mycoplasma or Ureaplasma (n⫽34); 4) mixed TLR2 (n⫽16) or 5) mixed TLR2 & TLR4-engaging microbes. (n⫽23). In vivo, IAI level was assessed by IL6 ELISA. An amniochorion explant system (term CS, n⫽4) was used to test the effect of prototype ligands in vitro. We generated dose response curves at 6 and 18h after challenge with ligands for TLR4 [LPS], TLR2 [Lipoteichoic acid (LTA), Pam3Cys (PAM), Mycoplasma FSL] alone or combined [LPS⫹PAM, LPS⫹FSL]. TNF␣, IL6, IL8 were measured in explant media and normalized to total protein. Tissue viability was confirmed by LDH release. RESULTS: In vivo, IAI with single GN had highest AF IL6 independent of GA (p⬍0.001). There was no difference in IL6 among single TLR2engaging spp. In contrast, mixed GP induced higher IL6 vs. single TLR2-engaging spp. (p⫽0.016). IAI with mixed GN & GP had higher IL6 vs. single and mixed GP (p⬍0.05). In vitro, LPS induced higher responses in all cytokines at all times and doses while FSL responses were lowest (p⬍0.001). Of all TLR2 ligands, LTA had the highest response (p⬍0.05). LPS⫹FSL produced a 3-fold increase in all cytokines over FSL alone (p⫽0.03). LPS⫹PAM induced a 20-fold increase over PAM alone (p⫽0.02) and a transient 4-fold increase over LPS alone at 6h (p⬍0.05). CONCLUSIONS: Both in vivo and in vitro, single TLR4-engaging bacteria generate highest inflammatory responses. Mixed infections lead to different IAI levels vs. single species alone.
77 Electronic fetal heart rate monitoring and infant mortality: a population-based study in the United States Han-Yang Chen1, Suneet Chauhan2, Cande Ananth3, Anthony Vintzileos4, Alfred Abuhamad2 1
Center for Urban Population Health, Milwaukee, WI, 2Eastern Virginia Medical School, Norfolk, VA, 3UMDNJ-Robert Wood Johnson Medical School, New Brunswick, NJ, 4Stony BrookWinthrop University Hospitals, Long Island, NY
OBJECTIVE: Evidence form randomized trials demonstrate an increase in intrapartum (IP) interventions due to electronic fetal heart rate monitoring (EFM) with virtually no improvement in perinatal mortality. Yet observational (“reality-based evidence,” Vintzileos Obstet Gynecol 2009) indicates a decrease in IP mortality. We examined the association between EFM during labor and corrected neonatal and infant mortality (CIM) in the US. STUDY DESIGN: We utilized the 2004 National Birth Cohortlinked birth and infant death data file restricted to singleton births with 1989 revision of the standard certificate of live birth, used by 41 states and DC. We excluded multiple gestations, anomalous newborns, foreign residents, gestational age (GA) ⬍24 weeks or ⱖ45 weeks, implausible birthweights, or if EFM use was unknown. Multivariable log-binomial regression models were fitted to estimate risk ratio (RR) and 99% confidence interval (CI) to evaluate the association between EFM and CIM, while adjusting for potential confounders: age, race, marital Status; education; smoking and, infant’s gender. RESULTS: There were 4,112,053 live births in 2004 and 47% (n⫽1,945,789) met the inclusion criteria; among them 88% had EFM. The risks of CIM were 3.0 per 1,000 births for those with and 3.8 per 1,000 births without EFM. The decrease in CIM varies by GA (Table I). CONCLUSIONS: In the U.S., the real life utilization of EFM was associated with decreased CIM. The magnitude of this decrease varied slightly by GA and it was exclusively driven by lower number of deaths during the early neonatal period.
sponses to infection. Gram positive (GP), Mycoplasma and Ureaplasma spp. primarily engage TLR2, while TLR4 recognizes Gram Supplement to JANUARY 2011 American Journal of Obstetrics & Gynecology
S43
Oral Concurrent Session 7
Infectious Disease/Intrapartum Fetal Assessment
Table I Association between EFM and corrected infant mortality: Estimated risk ratio (99% CI) Early (0-6 days) All
Late (7-27 days)
0.54 (0.46, 0.64) 0.90 (0.70, 1.15)
Postneonatal (28-364 days)
Infant (0-364 days)
0.94 (0.82, 1.08) 0.78 (0.71, 0.86)
...........................................................................................................................................................................................................
24-27 0.64 (0.53, 0.78) 0.93 (0.64, 1.34) 1.04 (0.71, 1.55) 0.79 (0.68, 0.91) ........................................................................................................................................................................................................... 28-31 0.52 (0.35, 0.78) 1.25 (0.59, 2.64) 1.05 (0.62, 1.77) 0.78 (0.59, 1.04) ........................................................................................................................................................................................................... 32-33 0.41 (0.19, 0.90) 0.99 (0.25, 3.88) 0.94 (0.45, 1.96) 0.72 (0.44, 1.17) ........................................................................................................................................................................................................... 34-36 0.51 (0.25, 1.03) 1.02 (0.45, 2.33) 1.12 (0.73, 1.73) 0.96 (0.69, 1.33) ........................................................................................................................................................................................................... 37-40 0.67 (0.43, 1.04) 0.74 (0.48, 1.15) 0.88 (0.74, 1.06) 0.84 (0.72, 0.98) ........................................................................................................................................................................................................... ⱖ41 0.65 (0.27, 1.58) 2.23 (0.48, 10.29) 1.05 (0.66, 1.66) 1.04 (0.70, 1.54) ........................................................................................................................................................................................................... Bold if significantly different
78 Evaluation of the inter-observer and intra-observer reliability of the NICHD 3-tier FHR interpretation system Sean Blackwell1, William A. Grobman2, Leah Antoniewicz1, Maria Hutchinson1, Cynthia Gyamfi3 1 University of Texas Health Science Center at Houston, Houston, TX, 2Northwestern University, Feinberg School of Medicine, Chicago, IL, 3Columbia University, New York, NY
OBJECTIVE: In 2008, a NICHD-sponsored workshop on Electronic Fetal Monitoring (EFM) recommended adoption of a 3-Tier system for the categorization of Fetal Heart Rate (FHR) patterns. Our purpose was to test the reliability of this new classification system. STUDY DESIGN: Three MFM specialists who participated in the NICHD workshop independently reviewed FHR tracings without clinical data. All FHR tracings were from women with term, singleton pregnancies who had known umbilical artery pH results. In order to maximize the potential for a variation in FHR patterns, we included
S44
www.AJOG.org
FHR tracings from women with umbilical pH ⬍ 7.00 (n⫽6), pH 7.00-7.10 (n⫽6), and pH ⬎ 7.10 (n⫽6). FHR tracings were divided into 15-20 minute epochs and random samples were selected. Reviewers were asked to classify the FHR tracing by NICHD category (Category I, II, or III) as well as describe the individual FHR parameters. Approximately 25% of FHR tracing were duplicated and reviewed in order to evaluate intra-observer agreement. Pre-defined criteria for agreement were used: kappa 0.0 - 0.20 (poor), 0.21-0.40 (fair), 0.410.6 (moderate), 0.61-0.8 (substantial), and 0.81-1.0 (almost perfect). RESULTS: Each reviewer evaluated 154 individual FHR segments. The inter-observer reliability for the 3 examiners by each FHR category and overall is described in the table. The intra-observer agreement ranged from substantial to perfect for the 3 examiners (kappa 0.741.0). There were no cases where a FHR tracing was classified as Category I by one examiner but Category III by another. Moderate interobserver agreement was present for individual FHR parameters: presence of FHR accelerations (kappa 0.58), early decelerations (kappa 0.49), variable decelerations (kappa 0.48), late decelerations (kappa 0.54), and FHR variability (overall kappa 0.57; absent 0.16, minimal 0.53, and moderate 0.69). CONCLUSIONS: Inter-observer agreement of 3-Tier FHR classification system was moderate for Categories I and II. Agreement for Category III tracings was poor mainly due to lack of agreement regarding absent variability. FHR category
Kappa
Level of Agreement
Category I
0.53
Moderate
Category II
0.46
Moderate
.......................................................................................................................................................................................... ..........................................................................................................................................................................................
Category III ⫺0.01 Poor .......................................................................................................................................................................................... Overall 0.47 Moderate ..........................................................................................................................................................................................
American Journal of Obstetrics & Gynecology Supplement to JANUARY 2011