Identification of Candidates for Progesterone: Why, Who, How, and When?

Identifying the Early Signs of Preterm Birth from U.S. Birth Records Using Machine Learning Techniques

Preterm birth (PTB) is the leading cause of infant mortality in the U.S. and globally. The goal of this study is to increase understanding of PTB risk factors that are present early in pregnancy by leveraging statistical and machine learning (ML) techniques on big data. The 2016 U.S. birth records were obtained and combined with two other area-level datasets, the Area Health Resources File and the County Health Ranking. Then, we applied logistic regression with elastic net regularization, random forest, and gradient boosting machines to study a cohort of 3.6 million singleton deliveries to identify generalizable PTB risk factors. The response variable is preterm birth, which includes spontaneous and indicated PTB, and we performed a binary classification. Our results show that the most important predictors of preterm birth are gestational and chronic hypertension, interval since last live birth, and history of a previous preterm birth, which explains 10.92, 5.98, and 5.63% of the predictive power, respectively. Parents' education is one of the influential variables in predicting PTB, explaining 7.89% of the predictive power. The relative importance of race declines when parents are more educated or have received adequate prenatal care. The gradient boosting machines outperformed with an AUC of 0.75 (sensitivity: 0.64, specificity: 0.73) for the validation dataset. In this study, we compare our results with seminal and most related studies to demonstrate the superiority of our results. The application of ML techniques improved the performance measures in the prediction of preterm birth. The results emphasize the importance of socioeconomic factors such as parental education as one of the most important indicators of preterm birth. More research is needed on these mechanisms through which socioeconomic factors affect biological responses.

Outcomes With Cerclage Alone Compared With Cerclage Plus 17α-Hydroxyprogesterone Caproate

OBJECTIVE

To examine the differences in perinatal outcomes among women with a prior preterm birth who received cerclage compared with cerclage plus 17α-hydroxyprogesterone caproate.

METHODS

Women with transvaginal cerclage placement and a prior delivery between 16 and 36 weeks of gestation were identified over a 10-year period (July 2002 to May 2012) in this retrospective cohort study. Exclusion criteria were delivery at another institution, abdominal cerclage, multiple gestations, and major fetal anomalies. Maternal demographics, gestational age at cerclage, gestational age at delivery, preterm prelabor rupture of membranes (PROM), and birth weight were compared between women with a cerclage and cerclage plus 17α-hydroxyprogesterone caproate. The primary outcome was delivery at less than 24 weeks of gestation.

RESULTS

Of the 411 women who had a cerclage, 260 met inclusion criteria. Of these, 171 received a cerclage alone and 89 received cerclage plus 17α-hydroxyprogesterone caproate. The two groups were not different with respect to maternal demographics and gestational age at cerclage. There was a significant difference among those who received indomethacin at the time of cerclage, betamethasone administration, and history of a loop electrosurgical excision procedure-cold knife cone and cerclage. Delivery at less than 24 weeks of gestation occurred in 6% of women receiving both 17α-hydroxyprogesterone caproate and cerclage compared with 16% in the cerclage only group (odds ratio [OR] 0.31, 95% confidence interval 0.10-0.78, P=.01). In the multivariate analysis controlling for indomethacin use, prior cerclage, and loop electrosurgical excision procedure-cold knife cone there was a 73% reduction in delivery in the combined treatment group compared with cerclage alone (adjusted OR 0.26, P=.02). A multivariant analysis was conducted with correction for indomethacin at the time of cerclage, prior cerclage, and loop electrosurgical excision procedure-cold knife cone and cerclage surgery. Even after controlling for significant variables, there remained a 73% reduction in delivery at less than 24 weeks of gestation in the cerclage plus 17α-hydroxyprogesterone caproate cohort (adjusted OR 0.26, P=.02).

CONCLUSION

Women receiving transvaginal cerclage plus 17α-hydroxyprogesterone caproate had a 69% relative reduction in delivery at less than 24 weeks of gestation when compared with women receiving cerclage alone. We found no difference in overall preterm delivery or preterm PROM. In this cohort, compared with cerclage alone, the likelihood of a viable neonate improves with both treatments.

Assessment of DHA on reducing early preterm birth: the ADORE randomized controlled trial protocol

BACKGROUND

Preterm birth contributes to 0.5 million deliveries in the United States (one of eight pregnancies) and poses a huge burden on public health with costs in the billions. Of particular concern is that the rate of earliest preterm birth (<34 weeks) (ePTB), which has decreased little since 1990 and has the greatest impact on the overall infant mortality, resulting in the greatest cost to society. Docosahexaenoic acid (DHA) supplementation provides a potential high yield, low risk strategy to reduce early preterm delivery in the US by up to 75%. We propose a Phase III Clinical Trial (randomized to low or high dose DHA, double-blinded) to examine the efficacy and safety of high dose DHA supplementation to reduce ePTB. We also plan for a secondary pregnancy efficacy analysis to determine if there is a subset of pregnancies most likely to benefit from DHA supplementation.

METHODS

Between 900 and 1200 pregnant women who are ≥ 18 years old and between 12 and 20 weeks gestation will be recruited from three trial experienced academic medical institutions. Participants will be randomly assigned to two daily capsules of algal oil (totaling 800 mg DHA) or soybean and corn oil (0 mg DHA). Both groups will receive a commercially available prenatal supplement containing 200 mg DHA. Therefore, the experimental group will receive 1000 mg DHA/d and the control group 200 mg DHA/d. We will then employ a novel Bayesian response adaptive randomization design that assigns more subjects to the "winning" group and potentially allows for substantially smaller sample size while providing a stronger conclusion regarding the most effective group. The study has an overall Type I error rate of 5% and a power of 90%. Participants are followed throughout pregnancy and delivery for safety and delivery outcomes.

DISCUSSION

We hypothesize that DHA will decrease the frequency of ePTB <34 weeks. Reducing ePTB is clinically important as these earliest preterm deliveries carry the highest risk of neonatal morbidity, as well as contribute significant stress for families and post a large societal burden.

TRIAL REGISTRATION

This trial was registered with ClinicalTrials.gov (identifier: NCT02626299 ) on December 8, 2015. Additional summary details may be found in Table 1.

Does adjunctive use of progesterone in women with cerclage improve prevention of preterm birth?. J Matern Fetal Neonatal Med 2017;31:202-208. [PMID: 28068860 DOI: 10.1080/14767058.2017.1280019]

OBJECTIVE

To evaluate outcomes among pregnancies with cerclage as compared to cerclage and adjunctive progesterone.

METHODS

A retrospective cohort study was performed from 1 October 2011-30 June 2015 including women with a singleton gestation with vaginal cerclage. Exclusion criteria included multiple gestations, simultaneous 17-alpha hydroxyprogesterone caproate (17-OHPC) and vaginal progesterone (vag-p) use, and patients lost to follow-up. Primary outcome was prevention of preterm birth less than 35 (PTB <35) weeks gestational age (GA).

RESULTS

One hundred thirty-six patients met inclusion criteria; 73 women had cerclage only, 53 had cerclage and 17-OHPC, 10 had cerclage and vag-p. GA at cerclage placement was similar across groups (p = 0.068). There was a difference in prevention of PTB <35 weeks GA among groups (p = 0.035) with a trend toward earlier delivery among patients with cerclage and vag-p. Rates of PTB <35 weeks in the cerclage (29%) and cerclage and 17-OHPC groups (34%) were similar (p = 0.533). The odds ratio for risk of PTB <35 weeks among women with cerclage and vag-p as compared to all other patients was 5.21 (95%CI: 1.3-21.2).

CONCLUSION

The combination of cerclage with intramuscular progesterone resulted in similar PTB prevention as compared to cerclage alone. There may be an association between cerclage, vaginal progesterone and higher rates of PTB which may be attributed to characteristics of the group rather than the therapies studied.

The safety of progestogen in the prevention of preterm birth: meta-analysis of neonatal mortality

BACKGROUND

The safety of preventive progestogen therapy for preterm birth remains to be established. This meta-analysis aimed to evaluate the effects of preventive progestogen therapy on neonatal mortality.

METHODS

Randomized controlled trials (RCTs) on the preventive use of progestogen therapy, published between October 1971 and November 2015, were identified by searching MEDLINE/PubMed, EMBASE, Scopus, ClinicalTrials.gov, Cochrane Library databases, CINAHL, POPLINE, and LILACS using "progesterone" and "preterm birth" as key terms. We conducted separate analyses according to the type of progestogen administered and plurality of the pregnancy.

RESULTS

Twenty-two RCTs provided data on 11,188 neonates. Preventive progestogen treatment in women with a history of preterm birth or short cervical length was not associated with increased risk of neonatal death compared to placebo in all analyzed progestogen types and pregnancy conditions. The pooled relative risks (95% confidence interval) of neonatal mortality were 0.69 (0.31-1.54) for vaginal progestogen in singleton pregnancies, 0.6 (0.33-1.09) for intramuscular progestogen in singleton pregnancies, 0.96 (0.51-1.8) for vaginal progestogen in multiple pregnancies, and 0.96 (0.49-1.9) for intramuscular progestogen in multiple pregnancies.

CONCLUSIONS

The results of this meta-analysis suggest that administration of preventive progestogen treatment to women at risk for preterm birth does not appear to negatively affect neonatal mortality in single or multiple pregnancies regardless of the route of administration.

Evaluating the impact of the laborist model of obstetric care on maternal and neonatal outcomes

BACKGROUND

The laborist model of obstetric care represents a change in care delivery with the potential of improving maternal and neonatal outcomes.

OBJECTIVE

We evaluated the effectiveness of the laborist model of care compared to the traditional model of obstetric care using specific maternal and neonatal outcome measures.

STUDY DESIGN

This is a population cohort study with laborist and nonlaborist hospitals matched 1:2 on delivery volume, geography, teaching status, and neonatal intensive care unit level using data from the National Perinatal Information Center/Quality Analytic Services database. A before-and-after study design with an untreated comparison group analyzed with the method of difference-in-differences was used to examine the impact of laborists on maternal and neonatal outcome measures within the 3 years after implementing the laborist system, after adjusting for secular trends, sociodemographic factors, and maternal medical conditions. The final outcome measures evaluated included cesarean delivery, chorioamnionitis, induction of labor, preterm birth, prolonged length of stay, Apgar at 5 minutes of <7, birth asphyxia, birth injury, birth trauma, and neonatal death.

RESULTS

We studied nearly 550,000 women from 24 hospitals (8 laborist and 16 nonlaborist hospitals) from 1998 through 2011. Implementation of laborists was associated with fewer labor inductions (adjusted odds ratio, 0.85; 95% confidence interval, 0.71-0.99) and decreased rate of preterm birth (adjusted odds ratio, 0.83; 95% confidence interval, 0.72-0.96) after controlling for confounders. Laborists did not impact the cesarean delivery rate, chorioamnionitis, or prolonged length of stay.

CONCLUSION

Implementation of the laborist model was associated with a significant reduction in labor induction rate and preterm birth without adversely affecting other outcomes.

17-Hydroxyprogesterone caproate (17OHP-C) coverage among eligible women delivering at 2 North Carolina hospitals in 2012 and 2013: A retrospective cohort study

BACKGROUND

Although a weekly injection of 17-hydroxyprogestone caproate is recommended for preventing recurrent preterm birth, clinical experience in North Carolina suggested that many eligible patients were not receiving the intervention.

OBJECTIVE

Our study sought to assess how well practices delivering at 2 major hospitals were doing in providing access to 17-hydroxyprogesterone caproate treatment for eligible patients.

STUDY DESIGN

This retrospective cohort analysis studied all deliveries occurring between January 1, 2012, and December 31, 2013, at 2 large hospitals in North Carolina. Women were included if they had a singleton pregnancy and history of a prior spontaneous preterm birth. We extracted demographic, payer, and medical information on each pregnancy, including whether women had been offered, accepted, and received 17-hydroxyprogesterone caproate. Our outcome of 17-hydroxyprogesterone caproate coverage was defined as documentation of ≥1 injection of the drug.

RESULTS

Over the 2-year study period, 1216 women with history of a prior preterm birth delivered at the 2 study hospitals, of which 627 were eligible for 17-hydroxyprogesterone caproate eligible after medical record review. Only 296 of the 627 eligible women (47%; 95% confidence interval, 43-51%) received ≥1 dose of the drug. In multivariable analysis, hospital of delivery, later presentation for prenatal care, fewer prenatal visits, later gestation of prior preterm birth, and having had a term delivery immediately before the index pregnancy were all associated with failed coverage. Among those women who were "covered," the median number of 17-hydroxyprogesterone caproate injections was 9 (interquartile range, 4-15), with 84 of 296 charts (28%) not having complete information on the number of doses.

CONCLUSION

Even under our liberal definition of coverage, less than half of eligible women received 17-hydroxyprogesterone caproate in this sample. Low overall use suggests that there is opportunity for improvement. Quality improvement strategies, including population-based measurement of 17-hydroxyprogesterone caproate coverage, are needed to fully implement this evidence-based intervention to decrease preterm birth.

Revisiting the cost-effectiveness of universal cervical length screening: importance of progesterone efficacy

BACKGROUND

Preterm birth (PTB) is a significant cause of neonatal morbidity and mortality. Studies have shown that vaginal progesterone therapy for women diagnosed with shortened cervical length can reduce the risk of PTB. However, published cost-effectiveness analyses of vaginal progesterone for short cervix have not considered an appropriate range of clinically important parameters.

OBJECTIVE

To evaluate the cost-effectiveness of universal cervical length screening in women without a history of spontaneous PTB, assuming that all women with shortened cervical length receive progesterone to reduce the likelihood of PTB.

STUDY DESIGN

A decision analysis model was developed to compare universal screening and no-screening strategies. The primary outcome was the cost-effectiveness ratio of both the strategies, defined as the estimated patient cost per quality-adjusted life-year (QALY) realized by the children. One-way sensitivity analyses were performed by varying progesterone efficacy to prevent PTB. A probabilistic sensitivity analysis was performed to address uncertainties in model parameter estimates.

RESULTS

In our base-case analysis, assuming that progesterone reduces the likelihood of PTB by 11%, the incremental cost-effectiveness ratio for screening was $158,000/QALY. Sensitivity analyses show that these results are highly sensitive to the presumed efficacy of progesterone to prevent PTB. In a 1-way sensitivity analysis, screening results in cost-saving if progesterone can reduce PTB by 36%. Additionally, for screening to be cost-effective at WTP=$60,000 in three clinical scenarios, progesterone therapy has to reduce PTB by 60%, 34% and 93%. Screening is never cost-saving in the worst-case scenario or when serial ultrasounds are employed, but could be cost-saving with a two-day hospitalization only if progesterone were 64% effective.

CONCLUSION

Cervical length screening and treatment with progesterone is a not a dominant, cost-effective strategy unless progesterone is more effective than has been suggested by available data for US women. Until future trials demonstrate greater progesterone efficacy, and effectiveness studies confirm a benefit from screening and treatment, the cost-effectiveness of universal cervical length screening in the United States remains questionable.

Racial disparity in previable birth

BACKGROUND

Extremely preterm birth of a live newborn before the limit of viability is rare but contributes uniformly to the infant mortality rate because essentially all cases result in neonatal death.

OBJECTIVE

The objective of the study was to quantify racial differences in previable birth and their contribution to infant mortality and to estimate the relative influence of factors associated with live birth occurring before the threshold of viability.

STUDY DESIGN

This was a population-based retrospective cohort of all live births in Ohio over a 7 year period, 2006-2012. Demographic, pregnancy, and delivery characteristics of previable live births at 16 0/7 to 22 6/7 weeks of gestation were compared with a referent group of live births at 37 0/7 to 42 6/7 weeks. Rates of birth at each week of gestation were compared between black and white mothers, and relative risk ratios were calculated. Logistic regression estimated the relative risk of factors associated with previable birth, with adjustment for concomitant risk factors.

RESULTS

Of 1,034,552 live births in Ohio during the study period, 2607 (0.25% of all live births) occurred during the previable period of 16-22 weeks. There is a significant racial disparity in the rate and relative risk of previable birth, with a 3- to 6-fold relative risk increase in black mothers at each week of previable gestational age. The incidence of previable birth for white mothers was 1.8 per 1000 and for black mothers, 6.9 per 1000. Factors most strongly associated with previable birth, presented as adjusted relative risk ratio (95% confidence interval [CI]), were maternal characteristics of black race adjusted relative risk 2.9 (95% CI, 2.6-3.2), age ≥ 35 years 1.3 (95% CI, 1.1-1.6), and unmarried 2.1 (95% CI, 1.8-2.3); fetal characteristics including congenital anomaly, 5.4 (95% CI, 3.4-8.1) and genetic disorder, 5.1 (95% CI, 2.5-10.1); and pregnancy characteristics including prior preterm birth 4.4 (95% CI, 3.7-5.2) and multifetal gestation, twin, 16.9 (95% CI, 14.4-19.8) or triplet, 65.4 (95% CI, 32.9-130.2). The majority, 80%, of previable births (16-22 weeks) were spontaneous in nature, compared with 73% in early preterm births (23-33 weeks), 72% in late preterm births (34-36 weeks), and 65% of term births (37-42 weeks) (P < .001). Previable births constituted approximately 28% of total infant mortalities in white newborns and 45% of infant mortalities in black infants in Ohio during the study period.

CONCLUSION

There is a significant racial disparity in previable preterm births, with black mothers incurring a 3- to 6-fold increased relative risk compared with white mothers, most of which are spontaneous in nature. This may explain much of the racial disparity in infant mortality because all live-born previable preterm births result in death. Focused efforts on the prevention of spontaneous previable preterm birth may help to reduce the racial disparity in infant mortality.

Assessment of adherence to guidelines for using progesterone to prevent recurrent preterm birth

OBJECTIVE

To assess if women with recurrent preterm birth had been offered, received, and adhered to progesterone supplementation guidelines and to ascertain reasons for failure to follow guidelines.

METHODS

Charts of infants who were products of recurrent spontaneous preterm birth were reviewed at four neonatal intensive care units of Akron Children's Hospital. Mothers of identified infants were interviewed and charts abstracted to determine: if progesterone was offered; acceptance of progesterone; compliance with progesterone; and reasons why progesterone was declined.

RESULTS

One hundred twenty-eight mothers with a recurrent spontaneous preterm birth were identified and 98 consented to participate. 62.2% (61/98) of the interviewed mothers reported that they were offered progesterone. Of the women offered progesterone, 82% (50/61) accepted treatment and 18.0% (11/61) declined. One woman who accepted progesterone did not receive it. Of the women who received progesterone, 18.4% (9/49) reported compliance failure. Of the women who did not receive progesterone, 75.5% (37/49) reported that they were not offered progesterone and 89.2% (33/37) of the women not offered progesterone reported that their care providers were aware of their prior preterm delivery.

CONCLUSIONS

Only 50% (49/98) of women who were candidates for progesterone received treatment. The main reason for women not receiving treatment was not being offered progesterone by their caregiver.

Human cervicovaginal fluid biomarkers to predict term and preterm labor

Preterm birth (PTB; birth before 37 completed weeks of gestation) remains the major cause of neonatal morbidity and mortality. The current generation of biomarkers predictive of PTB have limited utility. In pregnancy, the human cervicovaginal fluid (CVF) proteome is a reflection of the local biochemical milieu and is influenced by the physical changes occurring in the vagina, cervix and adjacent overlying fetal membranes. Term and preterm labor (PTL) share common pathways of cervical ripening, myometrial activation and fetal membranes rupture leading to birth. We therefore hypothesize that CVF biomarkers predictive of labor may be similar in both the term and preterm labor setting. In this review, we summarize some of the existing published literature as well as our team's breadth of work utilizing the CVF for the discovery and validation of putative CVF biomarkers predictive of human labor. Our team established an efficient method for collecting serial CVF samples for optimal 2-dimensional gel electrophoresis resolution and analysis. We first embarked on CVF biomarker discovery for the prediction of spontaneous onset of term labor using 2D-electrophoresis and solution array multiple analyte profiling. 2D-electrophoretic analyses were subsequently performed on CVF samples associated with PTB. Several proteins have been successfully validated and demonstrate that these biomarkers are associated with term and PTL and may be predictive of both term and PTL. In addition, the measurement of these putative biomarkers was found to be robust to the influences of vaginal microflora and/or semen. The future development of a multiple biomarker bed-side test would help improve the prediction of PTB and the clinical management of patients.