Preterm Labor.docx

Preterm labor General Information Description  regular uterine contractions occurring between 20 weeks and 36 weeks 6 days gestation and associated with cervical changes(1, 2, 3)  most common cause of antenatal hospitalization(1)  resulting preterm birth is leading cause of perinatal morbidity and mortality(1, 3)  periviable birth refers to delivery occurring between 20 weeks and 25 weeks 6 days gestation(4) Epidemiology Who is most affected  black women(2, 3)  women with multiple gestations(2, 3)  women using assisted reproductive technologies(2)  women of advanced maternal age(2)  maternal age < 16 years associated with increased risk for preterm labor, and risk increased with young gynecologic age (age ≤ 2 years from age at menarche)  based on cohort study  605 pregnant women evaluated  366 adolescents < 16 years old at time of last menstrual period  239 women aged 18-29 years at time of last menstrual period  36.3% of adolescents had young gynecologic age, defined as chronological age ≤ 2 years from age at menarche  factors associated with increased risk for preterm labor  age < 16 years (adjusted odds ratio [OR] 1.74, 95% CI 1.07-2.84)  age < 16 years with young gynecologic age (adjusted OR 2.64, 95% CI 1.23-5.65)  Reference - Ann Epidemiol 1997 Aug;7(6):400 Incidence/Prevalence  12.9 million preterm births estimated in 2005 (9.6% of all births worldwide), including  11.9% of births in Africa  10.6% of births in North America (excluding Mexico)  9.1% of births in Asia  8.1% of births in Latin American/Mexico and the Caribbean  6.4% of births in Australia/New Zealand  6.2% of births in Europe  Reference - Bull World Health Organ 2010 Jan;88(1):31 full-text  in United States(1, 2, 3)  about 12% of all births  percentage of births at < 34 weeks gestation decreased from 3.7% in 2006 to 3.4% in 2011  incidence varies by race in United States (data from 2005)(2)  about 18.4% for black women  about 12.1% for Hispanic women  about 11.7% for non-Hispanic white women  incidence increases with increasing number of fetuses(3)  about 50% of twin births reported to be preterm  about 90% of triplet births reported to be preterm Associated conditions



group B streptococcal infection(2)

Etiology and Pathogenesis Causes  often idiopathic - spontaneous preterm labor with intact membranes reported in about 50% of cases(2)  preterm premature rupture of membranes (PPROM) reported in about 25% of cases(2)  iatrogenic cause (elective induction of labor or cesarean delivery for medical indications) reported in about 25% of cases; indications include(2, 3)



hypertensive disorders of pregnancy

  

intrauterine growth restriction (IUGR)

nonreassuring fetal status placental abruption

Pathogenesis  decidual activation (paracrine signaling from fetus through amniotic fluid and across membranes to underlying maternal decidua and myometrium) triggers contractions(3)  in term labor, decidual activation occurs in late pregnancy when fetal pituitary adrenal axis reaches maturity  in preterm labor, decidual activation may be triggered prematurely via genetic and/or environmental factors, such as  choriodecidual inflammation  decidual hemorrhage  membrane rupture  uterine trauma or stretching  fetal signals of compromise  persistent contractions lead to increased cervical dilation and thinning over a period of hours before birth, resulting in active labor(3) Risk factors  prior premature birth (strongest risk factor)  cervical risk factors







 



  

  

incompetent cervix short cervix

(commonly defined as < 25 mm before 24 weeks gestation)

excisional treatment of cervical dysplasia

infections

  

bacterial vaginosis Chlamydia trachomatis other infections

drugs and toxins

    

tobacco alcohol

some medications (steroids, theophylline) pollutants

other toxins (such as diethylstilbestrol [DES] exposure in utero) psychological factors

  

depression stress or anxiety lower socioeconomic status

periodontal disease

(but evidence inconsistent)

fetal risk factors

  

multiple gestation congenital anomalies early fetal growth restriction

uterine and placental factors

     

placental abruption placenta previa

polyhydramnios uterine anomalies uterine leiomyoma

subchorionic hemorrhage (first trimester)

vaginal bleeding

subchorionic hematoma

other possible risk factors

       

trauma douching autoimmune conditions polycystic ovary syndrome

increased levels of inflammatory markers hereditary factor low fish intake

physically demanding work during pregnancy



     

< 20 years or > 40 years low maternal prepregnancy weight maternal weight gain < 0.27 kg (0.6 lbs)/week maternal age

infertility treatment time interval between pregnancies ethnicity

< 18 months or > 59 months

(black women)

factors not associated with increased risk of preterm labor

 

group B streptococcal (GBS) colonization prior abortion  see Risk factors for preterm labor and premature birth for details History and Physical History Chief concern (CC)  regular uterine contractions between 20 weeks and < 37 weeks gestation(1, 2, 3)  may be painful if contractions are sustained and precede clinically significant cervical preparation, such as contractions occurring after decidual hemorrhage(3)  often present as uterine contractions (typical in women with short cervix), characterized by(3)  mild but persistent pelvic pressure  cramps  increased vaginal discharge/mucus  occasional spotting over several day or weeks  watery discharge and/or vaginal bleeding may indicate true labor(2) History of present illness (HPI)  verify gestational age(2)  to determine likelihood of true labor, ask about characteristics of contractions, including(2)  strength  duration  frequency  associated vaginal bleeding or leaking of amniotic fluid  ask about recent uterine trauma(3) Past medical history (PMH)  ask about history of(2)  previous preterm labor  previous preterm delivery or pregnancy loss  infection (bacterial vaginosis, chlamydia, nongenital infections)  multiple gestation  shortened cervix (< 3 cm)  uterine anomalies  intrauterine growth retardation  placental abruption or placenta previa  cervical conization or loop electrosurgical excision procedure of the cervix  maternal periodontal infection  maternal abdominal surgery  maternal medical conditions including diabetes mellitus, hypertension, thyroid disease Social history (SH)  ask about history of(2)  smoking  substance abuse  stress  long work hours Physical General physical  assess fetal well-being(2)  assess for fever or other signs of infection(2,)  see Fetal monitoring during labor for details Pelvic  assess for membrane rupture(2,)  may be indicated by fluid leakage from cervical os on sterile speculum exam

  

obtain sample for culture and fetal fibronectin analysis nitrazine-positive reaction of fluid ferning of fluid  may need to reexamine after periods of recumbency to look for pooling of fluids if high suspicion of rupture of membranes  avoid digital exam if suspected rupture of membranes  see Preterm premature rupture of membranes (PPROM) for additional information  asses for factors associated with increased likelihood of true labor, including(2,)  contractions accompanied by  descent of presenting fetal part  progressive dilation and effacement of cervix  vaginal bleeding Diagnosis Making the diagnosis  diagnosis of preterm labor made clinically based on(1, 2)  regular uterine contractions accompanied by descent of presenting fetal part  progressive dilation and cervical effacement  confirmation of estimated gestational age < 37 weeks  true labor likely if ≥ 6 contractions/hour, cervical dilation ≥ 3 cm, ≥ 80% effacement, membrane rupture, and/or vaginal bleeding(2) Differential diagnosis  incorrect dating of pregnancy  normal preterm uterine activity (Braxton Hicks contractions) Testing overview



cervicovaginal fetal fibronectin (fFN)

  



cervical ultrasound

 



short cervix on ultrasound alone has low positive predictive value for preterm labor, and should not be used exclusively to direct management in symptomatic women (ACOG Level B) cervical length > 15 mm on transvaginal ultrasound may not predict delivery within 1 week in women with signs of preterm labor (level 2 [mid-level] evidence)

cervicovaginal and amniotic fluid biomarkers

 



positive fFN test results alone have low positive predictive value for preterm labor, and should not be used exclusively to direct management in symptomatic women (ACOG Level B) cervicovaginal fFN testing has limited ability to predict preterm birth within 7 days in symptomatic women (level 1 [likely reliable] evidence) negative fFN test may rule out delivery within 2 weeks in twin and singleton gestations in symptomatic women (level 2 [mid-level] evidence)

amniotic fluid assay of interleukin-6 ≥ 13.4 ng/dL or proteomic biomarkers each associated with increased likelihood of preterm birth in women with preterm labor but combination assay does not appear to be associated with improved detection (level 2 [mid-level] evidence) cervicovaginal interleukin-6 assay has lower sensitivity than fFN but similarly rules out preterm delivery within 14 days in women in preterm labor with low rates of preterm delivery (level 1 [likely reliable] evidence)

tests for infection

   

in preterm labor improves outcomes perform urinalysis and culture to assess for infection perform rectovaginal culture in unscreened patients to assess for group B Streptococcus, gonorrhea, and chlamydia obtain vaginal samples in symptomatic patients to assess for bacterial vaginosis and trichomonas Blood tests  white blood cell count > 12,000/mL before 28 weeks gestation and corticotrophin-releasing hormone > 684 pg/mL after 28 weeks may predict preterm labor in next 48 hours (level 2 [mid-level] evidence)  based on analysis of retrospective data and prospective cohort of women in threatened preterm labor  white blood cell (WBC) > 12,000/mL at 22-27 weeks gestation more accurate predictor of delivery within 48 hours, compared to corticotrophin-releasing hormone (CRH), cortisol, and maternal age  at 28-31 weeks gestation, CRH > 684 pg/mL more accurate predictor of delivery within 48 hours  Reference - Am J Obstet Gynecol 2008 Apr;198(4):468e1 Imaging studies  ultrasound-guided transabdominal amnioinfusion of indigo carmine (1 mL in 9 mL of normal saline) may confirm ruptured membranes in women with high suspicion of ruptured membranes despite negative findings on speculum exam(2)  cervical ultrasound used to determine cervical length no evidence to suggest that routine amniocentesis to check for infection





ultrasound screening for short cervix may be indicated in women with persistent pelvic pressure, cramps, spotting, and increased vaginal discharge(3)  short cervix on ultrasound alone has low positive predictive value for preterm labor, and should not be used exclusively to direct management in symptomatic women (ACOG Level B)(1)  cervical length > 15 mm on transvaginal ultrasound may not predict delivery within 1 week in women with signs of preterm labor (level 2 [mid-level] evidence)  based on systematic review limited by heterogeneity  systematic review of 28 cohort and case-control studies evaluating transvaginal ultrasound assessment of cervical length for prediction of preterm birth in women with singleton pregnancy, intact membranes, and preterm labor symptoms  7.1% of women delivered with 48 hours  11.1% of women delivered within 1 week from presentation  for prediction of preterm labor within 1 week, presentation < 34 weeks plus  cervical length > 15 mm associated with 96% pooled negative predictive value, results limited by significant heterogeneity  cervical length < 15 mm associated with 63% pooled negative predictive value, results limited by significant heterogeneity  Reference - Ultrasound Obstet Gynecol 2010 Jan;35(1):54 full-text  cervical length on ultrasound appears significantly associated with risk of preterm delivery, but predictive accuracy appears low in women with threatened preterm labor (level 2 [mid-level] evidence)  based on retrospective prognostic cohort study  1,077 pregnant women with preterm labor at < 34 weeks gestation had ultrasound determination of cervical length  correlation between cervical length and time to delivery was significant but weak (each additional 2 mm associated with 1 day increase in interval) (p < 0.001)  Reference - Obstet Gynecol 2013 Dec;122(6):1279  shortening of cervical length over repeated ultrasound measurements appears to have limited utility for predicting preterm birth (level 2 [mid-level] evidence)  based on systematic review with clinical heterogeneity  systematic review of 14 studies evaluating transvaginal sonographic cervical length over time for prediction of preterm birth in 4,398 pregnant women  7 studies provided data on women with singleton gestation and 8 studies provided data on women with twin gestations  time intervals between cervical length measurements and definition of positive result for shortening in cervical length between measurements varied across studies  pooled diagnostic performance of any cervical length shortening in singleton gestation  for preterm birth at < 37 weeks in analysis of 4 studies with 543 women  sensitivity 54% (95% CI 43%-65%)  specificity 84% (95% CI 80%-87%)  preterm birth at < 35 weeks in analysis of 5 studies with 2,979 women  sensitivity 65% (95% CI 57%-72%)  specificity 48% (95% CI 46%-50%)  pooled diagnostic performance of any cervical length shortening in twin gestation  for preterm birth at < 34 weeks in analysis of 7 studies with 852 women  sensitivity 47% (95% CI 39%-55%)  specificity 88% (95% CI 86%-91%)  for preterm birth at < 32 weeks in analysis of 4 studies with 644 women  sensitivity 61% (95% CI 49%-73%)  specificity 88% (95% CI 85%-91%)  consistent results for preterm birth < 30 weeks and at < 28 weeks in analysis of 3 studies with 454 women  Reference - Am J Obstet Gynecol 2015 Dec;213(6):789, commentary can be found in Evid Based Med 2016 Feb;21(1):40 fetal adrenal gland volume > 422 mm3 associated with increased likelihood of preterm birth within 5 days (level 2 [mid-level] evidence)  based on small prognostic cohort study without independent validation  126 women with singleton fetus had ultrasound evaluation of fetal adrenal gland volume  53 with signs or symptoms of preterm labor  73 controls  fetal adrenal gland volume successfully examined in 86.5%  diagnostic performance of corrected adrenal gland volume > 422 mm3/kg for predicting preterm birth within 5 days

   

sensitivity 92% specificity 99% positive likelihood ratio 93.5 negative likelihood ratio 0.08  Reference - Obstet Gynecol 2007 Apr;109(4):855, commentary can be found in Obstet Gynecol 2007 Jul;110(1):187  American College of Radiology (ACR) Appropriateness Criteria for assessment of gravid cervix can be found at ACR 2014 PDF, earlier version can be found at National Guideline Clearinghouse 2012 Apr 16:35156 Other diagnostic testing Fetal fibronectin (fFN)  positive result on fetal fibronectin (fFN) test has low positive predictive value for preterm labor, and should not be used exclusively to direct management in symptomatic women (ACOG Level B)(1)  negative results on fFN test may help determine which patients do not need tocolysis(1)  usual threshold indicating abnormal result for fFN test is > 50 ng/mL by enzyme-linked immunosorbent assay (BMJ 2002 Aug 10;325(7359):301 full-text)  possible causes of false-positive fFN test may include  sexual activity within 24 hours of sample collection  cervical examination within 24 hours of sample collection  cervical dilatation  uterine contractions  vaginal bleeding  preeclampsia  Reference - BMJ 2002 Aug 10;325(7359):301 full-text  cervicovaginal fFN testing has limited ability to predict preterm birth within 7 days in symptomatic women (level 1 [likely reliable] evidence)  based on systematic review  systematic review of 32 prognostic cohort studies evaluating fFN testing for predicting preterm birth within 7 days in 5,355 symptomatic patients  410 women (7.7%) delivered within 7 days of fFN testing  diagnostic performance of fFN testing for predicting preterm birth within 7 days  sensitivity 76.1%  specificity 81.9%  positive likelihood ratio 4.2  negative likelihood ratio 0.29  Reference - Obstet Gynecol 2009 Sep;114(3):631, commentary can be found in Obstet Gynecol 2010 Jan;115(1):186  negative fFN test may rule out delivery within 2 weeks in twin and singleton gestations in symptomatic women (level 2 [mid-level] evidence)  based on retrospective prognostic cohort study  429 singleton and 87 twin gestations with complaints of preterm labor had fFN testing  3.5% singletons and 28.7% twin pregnancies delivered before 34 weeks gestation Twins Singletons Sensitivity 71% 82% Specificity 74% 90% Positive predictive value 19% 17% Negative predictive value 97% 99% Abbreviation: fFN, fetal fibronectin test. Predictive Performance of fFN for Delivery within 14 days:  Reference - Obstet Gynecol 2007 May;109(5):1083  negative fFN may rule out delivery at < 30 weeks gestation in high-risk asymptomatic women with or without cervical cerclage (level 2 [mid-level] evidence)  based on retrospective prognostic cohort study without blinding  910 asymptomatic women at high risk of preterm birth had fFN testing at 23-27 weeks gestation  clinicians were not blinded to fFN test results which may have influenced management  diagnostic performance of negative fetal fibronectin test for determining likelihood of delivery at < 30 weeks gestation in women with vs. without cervical cerclage  specificity 77% vs. 90% (p < 0.0001)  sensitivity 60% vs. 78.6% (not significant)  negative predictive value > 98% in both groups (not significant)  Reference - BJOG 2009 May;116(6):799



positive fFN assay at 28 weeks (but not at 20-24 weeks) associated with increased risk of preterm birth in women with HIV infection (level 2 [mid-level] evidence)  based on subgroup analysis of randomized trial  2,353 African women with HIV infection randomized to antibiotics vs. placebo to prevent chorioamnionitis and reduce preterm birth and mother-to-child transmission of HIV had vaginal fluid collected prior to antibiotics (at 20-24 weeks) and after treatment at (28 weeks)  gestational age calculated by uterine size from pubic symphysis to uterine fundus (ultrasound not available)  comparing positive fFN (≥ 50 ng/mL) vs. negative (fFN < 50 ng/mL) at 28 weeks  preterm birth at < 32 weeks gestation in 10.8% vs. 1.9% (odds ratio [OR] 6.3, 95% CI 3.2-12.3)  preterm birth at < 37 weeks gestation in 38.7% vs. 22% (OR 2.3, 95% CI 1.5-3.3)  positive fFN at 20-24 weeks not associated with increased risk of preterm birth compared to fFN < 50 ng/mL  antibiotic treatment did not influence fFN levels  Reference - Obstet Gynecol 2007 Feb;109(2 Pt 1):392, correction can be found in Obstet Gynecol 2007 Oct;110(4):936  management based on fFN testing might reduce risk of preterm birth in women with singleton pregnancies and threatened preterm labor at 23-34 weeks gestation (level 2 [mid-level] evidence)  based on systematic review limited by clinical heterogeneity  systematic review of 6 randomized trials evaluating fFN testing in 546 women with singleton pregnancies and symptoms of preterm labor at 23-34 weeks gestation  trials evaluating fFN testing combined with sonographic cervical length excluded  definition of preterm labor varied across studies  comparing management based on fFN testing to control (test results blinded or no testing)  management based on fFN testing associated with nonsignificant decrease in preterm birth at < 37 weeks gestation (relative risk 0.72, 95% CI 0.52-1.01) in analysis of 5 trials with 434 women  no significant differences in  preterm birth at < 28, 32, or 34 weeks gestation in analyses of 4 trials with 357 women  delivery within 7 days in analysis of 4 days with 344 women  maternal hospitalization in analysis of 5 trials with 438 women  tocolysis in analysis of all trials  use of antenatal steroids in analysis of 5 trials with 438 women  neonatal respiratory distress syndrome in analysis of 2 trials with 148 neonates  neonatal intensive care admission in analysis of 2 trials with 129 neonates  Reference - Am J Obstet Gynecol 2016 Oct;215(4):431  similar results can be found in earlier Cochrane review evaluating fFN testing in 5 trials with 474 patients, including 4 trials in systematic review above (Cochrane Database Syst Rev 2008 Oct 8;(4):CD006843)  health technology assessment of rapid fetal fibronectin testing to predict preterm birth in women with symptoms of preterm labor can be found in Health Technol Assess 2013 Sep;17(40):1 PDF Cervicovaginal and amniotic fluid biomarkers  cervicovaginal interleukin-6 assay has lower sensitivity than fFN but similarly rules out preterm delivery within 14 days in women in preterm labor with low rates of preterm delivery (level 1 [likely reliable] evidence)  based on prognostic validation cohort study  660 cervicovaginal fluid samples from 552 women with singleton pregnancies at 24-34 weeks gestation were collected for fetal fibronectin assay  559 samples from women with symptoms of preterm labor  61 samples were from women without symptoms of preterm labor at time of collection  prevalence of preterm delivery  2.1% < 7 days after sample collection  4.7% < 14 days after sample collection Fetal Fibronectin > 50 mcg/LInterleukin-6 > 250 ng/L Preterm delivery within 7 days Sensitivity 93% 57% Specificity 86% 91% Positive predictive value 13% 12% Negative predictive value 100% 99% Preterm delivery within 14 days Sensitivity 65% 35% Specificity 87% 91% Positive predictive value 19% 16% Negative predictive value 98% 97% Predictive Performance of Assays for Preterm Delivery:

 

interleukin-6 (IL-6) to albumin ratio had poorer performance than interleukin-6 assay Reference - Clin Chem 2007 Aug;53(8):1534 full-text  combination of biomarkers  amniotic fluid assay of interleukin-6 ≥ 13.4 ng/dL or proteomic biomarkers each associated with increased likelihood of preterm birth in women with preterm labor, but combination assay does not appear to be associated with improved detection (level 2 [mid-level] evidence)  based on prognostic cohort study without independent validation  86 patients at 22-36 weeks gestation in preterm labor with intact membranes had assay of amniotic fluid interleukin6 and proteomic biomarkers (neutrophil defensins 1 and 2 and calgranulins A and C)  prevalence of intra-amniotic infection (defined as positive culture) 13.9%  interleukin-6 ≥ 13.4 ng/dL predicted preterm delivery < 37 weeks with  sensitivity 46%  specificity 93.8%  positive predictive value 92%  negative predictive value 52.6%  selected proteomic biomarkers predicted preterm delivery < 37 weeks with  sensitivity 30%  specificity 96.9%  positive predictive value 93.8%  negative predictive value 47%  combination assay did not improve prediction of preterm birth or neonatal morbidity  Reference - Am J Obstet Gynecol 2009 May;200(5):499e1  proteomic analysis of amniotic fluid may detect intra-amniotic inflammation and help determine time to delivery (level 2 [mid-level] evidence)  based on diagnostic cohort study without independent reference standard  169 consecutive women with singleton pregnancies admitted with preterm labor or preterm premature rupture of membranes had sample of amniotic fluid collected and analyzed  Mass Restricted (MR) score was calculated based on presence or absence of 4 protein biomarkers (neutrophil defensins 1 and 2 and calgranulins A and C)  no antenatal clinical tests considered gold standard for diagnosis of inflammation, and concurrent placental biopsy in utero not possible  amniocentesis-to-delivery interval mean 0.4 days for women with MR score 3-4 vs. 3.8 days for women with MR score 1-2 vs. 17 days for women with MR score 0 (p < 0.001)  Reference - PLoS Med 2007 Jan 16;4(1):e18 full-text  amniotic fluid analysis to assess for intra-amniotic infection  strategies to identify and treat risk factors in early pregnancy (such as amniocentesis for intra-amniotic infection) have not been shown to reduce rates of preterm birth(3)  low amniotic fluid glucose level does not appear to be a sensitive marker for detecting subclinical intrauterine infections in patients with preterm labor and intact membranes (level 2 [mid-level] evidence)  based on diagnostic case-control study  55 women in preterm labor and 58 women matched for gestational age who were not in preterm labor had assessment of amniotic fluid glucose levels, amniotic fluid cultures, and placental histologic characteristics  diagnostic performance of low amniotic fluid glucose level for detecting subclinical infection  sensitivity 41%-55%  specificity 94%-100%  Reference - Am J Obstet Gynecol 1994 Aug;171(2):365  amniotic fluid Gram stain may detect intra-amniotic infection in patients with preterm labor (level 2 [mid-level] evidence)  based on diagnostic case-control study  127 patients with preterm labor and 26 patients with preterm premature rupture of the membranes had amniocentesis and amniotic fluid culture  diagnostic performance of amniotic fluid Gram stain for detecting intra-amniotic infection  sensitivity 80%  specificity 91%  Reference - Am J Obstet Gynecol 1998 Sep;179(3 Pt 1):650 Other tests  vaginal fluid testing for assessment of rupture of membranes(2)  nitrazine test

 

tests fluid sample for alkalinity false-positive nitrazine results may occur in presence of blood, semen, alkaline urine, or vaginal infection  ferning  obtain vaginal fluid sample, place on glass slide and allow to air dry 10 minutes  development of fern-like crystalline pattern suggests presence of amniotic fluid  cervical mucus may also cause ferning  absence of fetal breathing movements or cervical length ≤ 15 mm each associated with increased likelihood of preterm birth within 2-7 days in pregnant women with labor symptoms (level 2 [mid-level] evidence)  based on systematic review of prognostic cohort studies with study-specific quality measures not reported  systematic review of 72 prognostic cohort studies evaluating fetal breathing movements (FBM),transvaginal sonographic cervical length measurements (TVS CL), or cervicovaginal fetal fibronectin (fFN) for determining likelihood of preterm birth within 2 or 7 days in 11,886 pregnant women with labor symptoms  pooled performance of tests for determining likelihood of preterm birth at 2 days and 7 days 2 Days 7 Days Positive Positive Likelihood Number of Likelihood Number of Test Sensitivity Specificity Ratio studies Sensitivity Specificity Ratio studies absence of 75% (95% CI 93% (95% CI 67% (95% CI 98% (95% CI FBM 57%-87%) 75%-98%) 10.4 4 43%-84%) 83%-100%) 31.6 7 TVS CL (cutoff ≤ 15 77% (95% CI 88% (95% CI 74% (95% CI 89% (95% CI mm) 54%-90%) 84%-91%) 6.4 9 58%-85%) 85%-92%) 6.8 15 62% (95% CI 81% (95% CI 75% (95% CI 79% (95% CI fFN 43%-78%) 74%-86%) 3.3 4 69%-80%) 76%-83%) 3.6 38  comparative performance for prediction of preterm birth  no significant differences among tests in sensitivity at 2 days and 7 days  FBM had significantly higher specificity than fFN and TVS CL at 2 days and 7 days  Reference - Am J Obstet Gynecol 2014 Jan;210(1):54.e1  ultrasound findings, amniotic fluid interleukin-6, and serum C-reactive protein may each be associated with increased likelihood of birth within 2-7 days in women with threatened preterm labor (level 2 [mid-level] evidence)  based on systematic review of mostly poor-quality studies  systematic review of 22 tests determining risk of spontaneous preterm birth among asymptomatic women in early pregnancy and symptomatic women with threatened preterm labor in later pregnancy  inadequate reporting of blinding in majority of trials  tests with likelihood ratio > 5 included  for predicting spontaneous preterm birth before 34 weeks gestation in asymptomatic women  cervical length measurement by ultrasound  cervicovaginal prolactin screening  fetal fibronectin screening  for predicting birth within 2-7 days in symptomatic women with threatened preterm labor  absence of fetal breathing movements  cervical length and funneling  amniotic fluid interleukin-6  serum C-reactive protein  for predicting birth before 34 or 37 weeks in symptomatic women with threatened preterm labor  matrix metalloprotease-9  amniotic fluid IL-6  cervicovaginal fetal fibronectin  cervicovaginal human chorionic gonadotrophin (hCG)  Reference - Health Technol Assess 2009 Sep;13(43):1  tests for infection(2)  perform urinalysis and culture to assess for infection  perform rectovaginal culture in unscreened patients to assess for group B Streptococcus, gonorrhea, and chlamydia  obtain vaginal samples in symptomatic patients to assess for bacterial vaginosis and trichomonas Treatment Treatment overview  reserve treatment for women with fetuses at a gestational age at which delaying delivery will benefit the neonate (Obstet Gynecol 2016 Jan;127(1):190)

 

obstetric interventions for pregnancies at risk of periviable delivery should focus on delaying delivery and improving neonatal outcomes should delivery occur (Obstet Gynecol 2015 Nov;126(5):e82) medications to delay delivery



tocolytics

      





tocolytics may prolong gestation for up to 48 hours, which may permit enough time for steroids to improve fetal lung maturity and for maternal transport to tertiary care facility generally, tocolytics not indicated for use before neonatal viability but may be considered for previable pregnancy to help inhibit contractions after an event known to cause preterm labor, such as intra-abdominal surgery tocolytics not usually indicated in women with preterm contractions without cervical change, especially women with cervical dilation < 2 cm maintenance treatment with tocolytic drugs does not improve perinatal outcomes and is not routinely recommended (ACOG Level A) tocolytic drugs not associated with clear reduction in perinatal or neonatal mortality or neonatal morbidity (RCOG Grade A) nifedipine, atosiban, and cyclo-oxygenase (COX) inhibitors have fewer types of side effects and less frequent side effects compared to beta agonists, but how they compare to one another is unclear (RCOG Grade A) calcium channel blockers are more effective than beta-mimetics for tocolysis and result in lower rates of neonatal morbidity (level 1 [likely reliable] evidence)

antibiotics



antibiotics indicated for specific infections if identified (such as group B Streptococcus), but no evidence of efficacy for prolonging gestation in women with preterm labor and intact membranes (ACOG Level A)  antibiotics may prolong pregnancy in women with preterm premature rupture of membranes but not in women with preterm labor and intact membranes (level 2 [mid-level] evidence)  in women in preterm labor with intact membranes, prophylactic antibiotics may reduce maternal infection but associated with increased neonatal mortality and cerebral palsy (level 2 [mid-level] evidence) and do not reduce preterm birth (level 1 [likely reliable] evidence) medications to improve pediatric outcomes



corticosteroids



American College of Obstetricians and Gynecologists recommendations single course of corticosteroids between 24 and 34 weeks gestation for women at risk of preterm delivery within 7 days  single course of corticosteroids before 32 weeks gestation in women with preterm premature rupture of membranes  dosing options include betamethasone 12 mg intramuscularly every 24 hours for 2 doses, or dexamethasone 6 mg intramuscularly every 12 hours for 4 doses  antenatal steroids for women at risk of preterm birth associated with reduced neonatal mortality and respiratory distress syndrome (RDS) (level 2 [mid-level] evidence)  repeat doses of prenatal corticosteroids reduce RDS in infants at risk of preterm birth, but no significant differences in childhood outcomes at early childhood follow-up (18 months to 2 years corrected age) (level 1 [likely reliable] evidence)  prenatal dexamethasone might reduce risk of intraventricular hemorrhage compared to prenatal betamethasone (level 3 [lacking direct] evidence), but no significant differences in neonatal mortality or most major neonatal morbidities (level 2 [mid-level] evidence)  prenatal dexamethasone exposure may be associated with higher risk for neurodevelopmental abnormalities than prenatal betamethasone exposure (level 2 [mid-level] evidence)  antenatal magnesium sulfate  if birth anticipated < 32 weeks, administer magnesium sulfate to reduce severity and risk of cerebral palsy in neonate (ACOG Level A)  World Health Organization recommends magnesium sulfate in women at risk of imminent preterm birth < 32 weeks gestational age for prevention of cerebral palsy (Strong recommendation based on moderate-quality evidence)  recommended dose 4-6 g IV bolus for 20 minutes followed by 1-2 g/hour (max 3 g/hour)  antenatal magnesium sulfate in women at risk for preterm birth decreases risk of cerebral palsy and substantial gross motor dysfunction (level 1 [likely reliable] evidence)  addition of thyrotropin-releasing hormone to steroids in women at risk of very preterm delivery does not reduce mortality or adverse fetal outcomes, and causes maternal side effects (level 1 [likely reliable] evidence)  IV hydration does not appear to reduce preterm delivery before 37 weeks for women in preterm labor  management in hospital or at home after initial treatment of preterm labor associated with similar preterm birth rates (level 2 [mid-level] evidence) Treatment setting  antenatal transport to hospital with advanced neonatal and maternal care is recommended based on anticipated neonatal or maternal complications associated with periviable birth (SMFM/ACOG Best practice)(4)





management in hospital or at home after initial treatment of preterm labor associated with similar preterm birth rates (level 2 [mid-level] evidence)  based on randomized trial with incomplete blinding of caregivers  250 women > 18 years old at 20-35 weeks gestation experiencing first admission to hospital for first episode of preterm labor were randomized to home care (early discharge with nurse home visits) vs. hospital care after treatment for acute episode of premature labor  all women had cervical dilation of < 4 cm with < 80% effacement, lived within 50 km (31.1 miles) of hospital, and had no prior preterm delivery  women in hospital group were discharged earlier than they might otherwise have been because their physicians were aware that there were no major emergency readmissions or deliveries in the home group  no significant differences between groups in  mean gestational age at delivery  mean birth weight  preterm delivery rate  mean duration of neonatal hospital stay  mean duration of neonatal intensive care unit stay  Reference - CMAJ 2001 Apr 3;164(7):985 full-text  hospitalization may not be associated with reduced rate of preterm birth (level 2 [mid-level] evidence)  based on randomized trial with early termination and inadequate statistical power  101 women with arrested preterm labor and intact fetal membranes between 24 and 33 4/7 weeks gestation were randomized to home or hospital management following course of dexamethasone  study terminated before planned interim analysis due to no apparent differences between groups after 6 years  sample size at study termination had 60% power to detect 30% relative reduction in delivery at ≥ 36 weeks  no significant differences between groups in delivery at ≥ 36 weeks gestation  hospitalization associated with higher rate of infants with respiratory distress requiring mechanical ventilation for at least 24 hours compared to home monitoring (13% vs. 2%, p = 0.04)  Reference - Obstet Gynecol 2005 Jul;106(1):14, editorial can be found in Obstet Gynecol 2005 Jul;106(1):3  delivery at hospital with high-level neonatal intensive care unit associated with fewer in-hospital deaths among premature neonates (level 2 [mid-level] evidence)  based on retrospective cohort study  1,328,132 hospital deliveries of premature neonates delivered at 23-37 weeks gestation in hospitals in Pennsylvania, California, and Missouri evaluated  high-level hospital defined as level III facility with mean annual delivery of ≥ 50 very-low-birth-weight infants  delivery at hospital with high-level neonatal intensive care unit associated with  7.8 fewer in-hospital deaths per 1,000 deliveries in Pennsylvania, United States  2.7 fewer in-hospital deaths per 1,000 deliveries in California, United States  12.6 fewer in-hospital deaths per 1,000 deliveries in Missouri, United States  Reference - Pediatrics 2012 Aug;130(2):270 full-text Fluid and electrolytes  routine IV hydration has not been shown to reduce risk for preterm birth (ACOG Level B)(1)  IV hydration does not appear to reduce preterm delivery before 37 weeks for women in preterm labor (level 2 [midlevel] evidence)  based on Cochrane review with limited evidence from trials that lacked blinding  systematic review of 2 randomized trials evaluating oral or IV hydration in 228 women with preterm labor  trials compared IV hydration (bolus of 500 mL of crystalloid over 20 minutes followed by continuous infusion of 200 mL/hour) vs. bed rest alone in 228 women meeting inclusion criteria  no significant differences in  preterm delivery before 37 weeks gestation in analysis of 2 trials with 228 women  neonatal intensive care unit admission in 1 trial with 118 patients  no trials evaluated oral hydration  Reference - Cochrane Database Syst Rev 2013 Nov 4;(11):CD003096 Counseling  for threatened and imminent periviable birth(4)  use best estimate of gestational age to guide counseling and decision making for periviable births  prenatal and postnatal counseling regarding anticipated short-term and long-term neonatal outcomes should consider (SMFM/ACOG Best practice)  anticipated gestational age at delivery  other variables that may impact likelihood of survival and adverse outcomes, such as

     

fetal sex multiple gestation presences of suspected major malformations antenatal corticosteroid administration birth weight response to initial resuscitation  formulate antenatal plan of care with parents focusing on optimizing chance of survival or minimizing suffering Medications Medications to delay delivery Tocolytics  reserve tocolytic therapy for women with fetuses that would benefit from 48-hour delay in delivery(1)  administer first-line tocolytic treatment with beta-adrenergic agonist therapy, calcium channel blockers, or nonsteroidal antiinflammatory drugs (NSAIDs) for short-term prolongation of pregnancy to allow time for administrations of steroids (ACOG Level A)(1)  generally, tocolytics not indicated for use before neonatal viability but may be considered for previable pregnancy to help inhibit contractions after an event known to cause preterm labor, such as intra-abdominal surgery(1)  recommendations for tocolytics in preterm labor for threatened and imminent periviable birth to allow for antenatal corticosteroid administration based on best estimate of gestational age(4)  recommended for pregnancies ≥ 24 weeks gestation (SMFM/ACOG Grade 1B)  consider for pregnancies 23 weeks gestation (SMFM/ACOG Grade 2B)  not recommended for pregnancies > 20 weeks gestation but < 23 weeks gestation (SMFM/ACOG Grade 1A)  tocolytics not usually indicated in women with preterm contractions without cervical change, especially women with cervical dilation < 2 cm(1)  prophylactic tocolytics may not reduce risk for preterm birth or improve neonatal outcomes in women with multiple gestations, and may be associated with increased risk for maternal complications (such as pulmonary edema)(1)  American College of Obstetricians and Gynecologists (ACOG) recommendations on role of tocolysis  tocolytic agents may be used for short-term prolongation of pregnancy (up to 48 hours) to allow for administration of antenatal steroids (ACOG Level A)  generally, tocolytics not indicated for use before neonatal viability but may be considered for previable pregnancy to help inhibit contractions after an event known to cause preterm labor, such as intra-abdominal surgery  maintenance treatment with tocolytic drugs does not improve perinatal outcomes and is not routinely recommended (ACOG Level A)  tocolytic drugs not associated with clear reduction in perinatal or neonatal mortality or neonatal morbidity (RCOG Grade A)  World Health Organization does not recommend tocolytic drugs to improve newborn outcomes in women at risk of imminent preterm birth (WHO Conditional recommendation, Very low-quality evidence) (WHO 2015 PDF)  contraindications to tocolysis  tocolytics not usually indicated in women with preterm contractions without cervical change, especially women with cervical dilation < 2 cm  other contraindications include  intrauterine fetal demise  lethal fetal anomaly  nonreassuring fetal status  severe preeclampsia or eclampsia  maternal bleeding with hemodynamic instability  chorioamnionitis  preterm premature rupture of membranes (PPROM) (tocolysis may be considered for purpose of maternal transport and/or steroid administration)  medication-specific contraindications to tocolysis  choice of tocolytic agent  ACOG recommendations for choice of tocolytic agent  first-line agents include any of (ACOG Level A)  beta-adrenergic agonist therapy  calcium channel blockers (nifedipine)  nonsteroidal anti-inflammatory drugs (NSAIDs)  magnesium sulfate - The American College of Obstetricians and Gynecologists (ACOG) and The Society for Maternal-Fetal Medicine support use of magnesium sulfate for short-term prolongation of pregnancy (up to 48 hours) to allow for administration of antenatal steroids in pregnant women at risk of preterm delivery within 7 days  nifedipine, atosiban, and cyclooxygenase (COX) inhibitors have fewer types of side effects and less frequent side effects compared to beta agonists, but how they compare to one another is unclear (RCOG Grade A)



may decrease risk of preterm birth and admission to neonatal intensive care unit compared to betamimetics in women with preterm labor (level 2 [mid-level] evidence)  NSAIDs appear to prevent preterm birth and appear more effective than beta-mimetics (level 2 [mid-level] evidence)  beta-mimetics delay delivery but do not decrease respiratory distress syndrome or mortality outcomes (level 3 [lacking direct] evidence) and use limited by adverse effects  selected options include  nifedipine 20 mg orally, then 10-20 mg 3-4 times daily for up to 48 hours  NSAIDs (indomethacin)  50 mg rectal suppository or 50-100 mg orally as loading dose  25-50 mg orally every 4 hours for 48 hours subsequently  terbutaline 0.25 mg subcutaneously every 20 minutes to every 3 hours based on maternal uterine activity and pulse  ritodrine initially 0.1 mg/minute IV, increased 0.5 mg/minute every 10 minutes until labor controlled, maximum 0.35 mg/minute  once labor inhibited, reduce dose by 0.5 mg/minute every 30 minutes to lowest effective dose  duration of infusion arbitrary, but usually 12-14 hours  check potassium and glucose in 4 hours  atosiban 6.75 mg over 1 minute as initial bolus followed by 18 mg/hour for 3 hours, then 6 mg/hour for up to 45 hours (maximum total dose 330 mg)  magnesium sulfate 4-6 g IV bolus over 15-20 minutes, then 2 g/hour IV (may be increased to 4-5 g/hour as needed if no significant adverse effects or oliguria)  see Tocolytics for treatment of preterm labor for details Antibiotics  World Health Organization recommendations for antibiotics in women with preterm labor  does not give routine antibiotics if amniotic membranes are intact (WHO Strong recommendation, Moderate-quality evidence)  administer antibiotics if membranes have ruptured (WHO Strong recommendation, Moderate-quality evidence)  erythromycin is antibiotic of choice (WHO Conditional recommendation, Moderate-quality evidence)  do not use amoxicillin/clavulanic acid (WHO Strong recommendation, Moderate-quality evidence)  Reference - WHO 2015 PDF  antibiotics indicated for specific infections if identified (such as group B Streptococcus [GBS]), but no evidence of efficacy for prolonging gestation in women with preterm labor and intact membranes (ACOG Level A)(1, 2)  indications for GBS prophylaxis  women with GBS bacteriuria at any time during current pregnancy  intrapartum prophylaxis for women with positive GBS colonization  women with prior infant with invasive GBS disease  women with culture results unknown if any of  < 37 weeks gestation  duration of membrane rupture ≥ 18 hours  temperature ≥ 100.4 degrees F (38 degrees C)  threatened preterm delivery until culture results known  see Group B streptococcal infection in infants less than 3 months old for details  periviable birth  recommendations for antibiotics to prolong latency during expectant management of preterm premature rupture of membranes if delivery not imminent for threatened periviable birth based on best estimate of gestational age(4)  recommended for pregnancies ≥ 24 weeks gestation (SMFM/ACOG Grade 1B)  consider for pregnancies 23 weeks gestation (SMFM/ACOG Grade 2B)  consider for pregnancies > 20 weeks gestation but < 23 weeks gestation (SMFM/ACOG Grade 2C)  intrapartum antibiotics for GBS prophylaxis for threatened and imminent periviable birth based on best estimate of gestational age(4)  recommended for pregnancies ≥ 24 weeks gestation (SMFM/ACOG Grade 1B)  consider for pregnancies 23 weeks gestation (SMFM/ACOG Grade 2B)  not recommended for pregnancies > 20 weeks gestation but < 23 weeks gestation (SMFM/ACOG Grade 1A)  evidence  antibiotics may prolong pregnancy in women with preterm premature rupture of membranes but not in women with preterm labor and intact membranes (level 2 [mid-level] evidence)  based on systematic review without assessment of trial quality  systematic review of 22 randomized trials evaluating antibiotics for women at 22-34 weeks gestational age presenting with preterm premature rupture of membrane (PPROM) or preterm labor  in women with PPROM antibiotics associated with calcium channel blockers





slightly increased latency period (weighted mean difference 0.33 days, 95% CI 0.17-0.5 days) in analysis of 5 trials with 4,435 women  reduced rate of delivery within 48 hours (odds ratio [OR] 0.65, 95% CI 0.57-0.74) in analysis of 5 trials with 4,974 women  reduced rate of delivery within 7 days (OR 0.67, 95% CI 0.59-0.76) in analysis of 5 trials with 5,001 women, results limited by significant heterogeneity  reduced rate of clinical neonatal sepsis (OR 0.71, 95% CI 0.52-0.97) in analysis of 9 trials with 1,214 women  in women with preterm labor (intact membranes)  no significant difference in latency period (weighted mean difference 0.21 days, 95% CI -1.36 days to +1.78 days) in analysis of 6 trials with 6,204 women  no significant difference in rates of delivery within 48 hours (OR 1.07, 95% CI 0.88-1.3) in analysis of 2 trials with 6,037 women  no significant difference in rates of delivery within 7 days (OR 1.03, 95% CI 0.88-1.21) in analysis of 5 trials with 6,267 women  no significant difference in neonatal mortality (OR 0.98, 95% CI 0.69-1.39) in analysis of 9 trials with 6,686 women  significant reduction in clinical neonatal sepsis (OR 0.43, 95% CI 0.27-0.68) in analysis of 9 trials with 1,004 women, but results limited by heterogeneity  Reference - Am J Obstet Gynecol 2008 Dec;199(6):620e1, editorial can be found in Am J Obstet Gynecol 2008 Dec;199(6):583 in women in preterm labor with intact membranes, prophylactic antibiotics may reduce maternal infection but associated with increased neonatal mortality and cerebral palsy (level 2 [mid-level] evidence) and do not reduce preterm birth (level 1 [likely reliable] evidence)  based on Cochrane review with wide confidence intervals or possible publication bias or small-study effects for many outcomes  systematic review of 14 randomized trials comparing prophylactic antibiotics to placebo or no treatment in 7,837 women in preterm labor with intact membranes  most data came from 1 large trial (summarized below)  no significant differences in most outcomes in overall meta-analyses  preterm birth (relative risk [RR] 0.98, 95% CI 0.92-1.05)  perinatal mortality (combination of stillbirth and neonatal mortality) (RR 1.22, 95% CI 0.88-1.69)  stillbirth (RR 0.73, 95% CI 0.43-1.26)  infant mortality (RR 1.06, 95% CI 0.68-1.67)  birth weight (mean difference 58.38 g, 95% CI -26.24 g to +143 g)  admission to neonatal intensive care or special care (RR 0.82, 95% CI 0.62-1.1)  prophylactic antibiotics associated with reduced maternal infection in analysis of 10 trials with 7,371 women  RR 0.74 (95% CI 0.63-0.86)  NNT 25-65 with maternal infection in 11% of controls  asymmetrical funnel plot suggesting possible publication bias or small-study effects  increased neonatal mortality associated with  any prophylactic antibiotics in analysis of 9 trials with 7,248 infants  RR 1.57 (95% CI 1.03-2.4)  NNH 59-2,777 with neonatal mortality in 1.2% of controls  macrolide antibiotics in analysis of 3 trials with 6,684 infants (RR 1.52, 95% CI 1.05-2.19)  beta-lactam antibiotics in analysis of 7 trials with 7,053 infants (RR 1.51, 95% CI 1.06-2.15)  in single trial with 3,173 infants followed at age 7 years  macrolide antibiotics associated with increased risk of functional impairment (RR 1.11, 95% CI 1.01-1.2) and cerebral palsy (RR 1.9, 95% CI 1.2-3.01)  beta-lactam antibiotics associated with increased risk of cerebral palsy (RR 1.67, 95% CI 1.06-2.61)  Reference - Cochrane Database Syst Rev 2013 Dec 5;(12):CD000246  antibiotics do not improve neonatal mortality or morbidity when taken by women in spontaneous preterm labor without evidence of clinical infection (level 1 [likely reliable] evidence) and may be associated with increased risk for cerebral palsy in children at age 7 years (level 2 [mid-level] evidence)  based on largest randomized trial included in Cochrane review above with limited follow-up at 7 years  6,295 women in spontaneous preterm labor with intact membranes and no evidence of clinical infection were randomized to 1 of 4 treatments (orally 4 times daily for 10 days or until delivery)  erythromycin 250 mg  co-amoxiclav (amoxicillin 250 mg/clavulanic acid 125 mg)  erythromycin plus co-amoxiclav

 



placebo comparing treatment groups, no significant differences in combined outcome (neonatal death, chronic lung disease, or major cerebral abnormality on ultrasonography before hospital discharge) Reference - ORACLE II trial (Lancet 2001 Mar 31;357(9261):989), editorial can be found in Lancet 2001 Mar 31;357(9261):973, commentary can be found in ACP J Club 2001 Sep-Oct;135(2):70, Lancet 2001 Oct 6;358(9288):1184, Lancet 2001 Nov 17;358(9294):1728



in follow-up study of children born to 4,221 women completing ORACLE II trial with significant loss to follow-up outcomes available for 3,196 children (71%) cerebral palsy occurred in  3.3% with vs. 1.7% without erythromycin (p < 0.05, NNH 63)  3.2% with vs. 1.9% without co-amoxiclav (p < 0.05, NNH 77)  functional impairment occurred in  42.3% with vs. 38.3% without erythromycin (p < 0.05, NNH 25)  co-amoxiclav (with or without erythromycin) had no effect on functional impairment  no significant differences in  mortality  other medical conditions including fits/seizures, hydrocephalus with shunt, wheezing, diabetes, bowel disorders, and attention deficit disorder  hospital admissions  medication for asthma  behavioral patterns  educational attainment compared to national curriculum test results  Reference - Lancet 2008 Oct 11;372(9646):1319, editorial can be found in Lancet 2008 Oct 11;372(9646):1276, commentary can be found in Lancet 2009 Jan 3;373(9657):25

 

Heparin limited evidence that heparin reduces preterm birth rates in women at risk of complications due to placental dysfunction (level 2 [mid-level] evidence)  based on systematic review of low-quality trials  systematic review of 1 randomized trial and 1 cohort study evaluating heparin during pregnancy for women at risk of complications including preterm birth < 37 weeks gestation  randomized trial had unclear reporting of blinding, randomization, and allocation concealment  in randomized trial with 107 women, heparin was not associated with reduction in preterm birth < 37 weeks gestation or low birth weight  Reference - Acta Obstet Gynecol Scand 2008;87(8):804 Progesterones  reserve antenatal progesterone prophylaxis for women with previous spontaneous birth at < 37 weeks gestation(2, 3)  vaginal progesterone may not reduce preterm delivery in women with singleton pregnancies and threatened preterm labor (level 2 [mid-level] evidence)  based on randomized trial with low compliance  385 women with singleton pregnancies and threatened preterm labor at 24-33 gestational weeks were treated with acute tocolysis and randomized within 48 hours to self-administered progesterone 200 mg vaginally daily vs. placebo until 36 6/7 weeks gestation or preterm delivery  all women had intact membranes at baseline  trial had planned to recruit 626 women but was terminated early due to futility following prespecified interim analysis  42% in each group were noncompliant (defined as < 80% use of prescribed medication)  comparing progesterone vs. placebo  delivery at < 37 gestational weeks in 42.5% vs. 35.5% (not significant)  delivery at < 34 gestational weeks in 19.7% vs. 12.9% (p = 0.1, not significant)  delivery at < 32 gestational weeks in 12.9% vs. 9.7% (not significant)  neonatal morbidity in 22.8% vs. 18.8% (not significant)  no significant differences in duration of tocolysis, hospitalization, or recurrence of preterm labor  Reference - BJOG 2015 Jan;122(1):80  progesterone may reduce preterm delivery at < 37 gestational weeks for women with threatened preterm labor and intact membranes (level 2 [mid-level] evidence)  based on Cochrane review with limited evidence  systematic review of 3 randomized trials and 1 quasi-randomized trial evaluating progestational agents for women with threatened or established preterm labor with intact membranes in 307 patients  progesterone associated with reduced risk of preterm delivery at < 37 gestational weeks in analysis of 2 trials with 104 patients with threatened preterm labor



 

risk ratio 0.33 (95% CI 0.17-0.67) NNT 3-7 with preterm delivery in 46.2% of controls  Reference - Cochrane Database Syst Rev 2010 Jan 20;(1):CD006770 Medications to improve pediatric outcomes Corticosteroids Corticosteroids for gestations up to 34 completed weeks  recommendations for antenatal steroids in pregnancies up to 34 completed weeks gestation  American College of Obstetricians and Gynecologists (ACOG) recommendations for antenatal corticosteroid therapy for fetal maturation  single course of corticosteroids between 24 and 34 weeks gestation recommended for women at risk of preterm delivery within 7 days (ACOG Level A)  consider single course of corticosteroids for pregnant women starting at 23 weeks gestation who are at increased risk of preterm delivery within 7 days (ACOG Level A)  recommended corticosteroid regimen includes 1 of  betamethasone 12 mg intramuscularly every 24 hours for 2 doses, or  dexamethasone 6 mg intramuscularly every 12 hours for 4 doses  consider single rescue course if antecedent treatment given ≥ 7 days ago and woman remains at risk of preterm birth before 34 weeks gestation (ACOG Level B)  regularly scheduled repeat courses or multiple courses (> 2) not recommended  References - Obstet Gynecol 2016 Jan;127(1):190  American College of Obstetricians and Gynecologists and Society for Maternal and Fetal Medicine consensus guidelines on antenatal corticosteroids for threatened and imminent periviable birth based on best estimate of gestational age  recommended for pregnancies ≥ 24 weeks gestation (SMFM/ACOG Grade 1B)  consider for pregnancies 23 weeks gestation (SMFM/ACOG Grade 2B)  not recommended for pregnancies > 20 weeks gestation but < 23 weeks gestation (SMFM/ACOG Grade 1A)  Reference - Obstet Gynecol 2015 Nov;126(5):e82  World Health Organization (WHO) recommendations on antenatal corticosteroid therapy for women at risk of preterm birth from 24-34 weeks gestational age  indicated if (WHO Strong recommendation, Moderate-quality evidence for newborn outcomes; WHO Strong recommendation, Low-quality evidence for maternal outcomes)  gestational age assessment can be done accurately  preterm birth is imminent (within 7 days of starting treatment, including within first 24 hours)  there is no clinical evidence of maternal infection, even if membranes have ruptured  adequate childbirth care is available (including capacity to recognize and safely manage preterm labour and birth)  preterm neonate can receive adequate care if needed (including resuscitation, thermal care, feeding support, infection treatment and safe oxygen use)  administer intramuscular dexamethasone or betamethasone 24 mg in divided doses (WHO Strong recommendation, Lowquality evidence)  administer even if  single or multiple birth is anticipated. (WHO Strong recommendation, Low-quality evidence)  mother has hypertensive disorder (WHO Strong recommendation, Moderate-quality evidence for newborn outcomes; WHO Strong recommendation, Low-quality evidence for maternal outcomes) or pre-gestational and gestational diabetes (WHO Strong recommendation, Very low-quality evidence)  intrauterine growth restriction is present (WHO Strong recommendation, Very low-quality evidence)  do not administer if  chorioamnionitis is present (WHO Conditional recommendation, Very low-quality evidence)  planned cesarean section done at 34-36 weeks gestational age (WHO Conditional recommendation, Very low-quality evidence)  if delivery has not occurred repeat dose in 7 days if clinical assessment demonstrates high risk of preterm birth within 7 days. (WHO Conditional recommendation, Moderate-quality evidence for newborn outcomes: WHO Conditional recommendation, low-quality evidence for maternal outcomes)  Reference - WHO 2015 PDF  European consensus guidelines on management of neonatal respiratory distress syndrome (RDS) in preterm infants  prenatal steroids given to women with anticipated preterm delivery reduce risk of neonatal death  optimal treatment-to-delivery interval > 24 hours and < 7 days after starting steroid treatment  offer single course of prenatal corticosteroids to all women at risk of preterm delivery from about 23 weeks to 34 completed weeks gestation (European consensus Grade A)





consider short-term use of tocolytic drugs to allow completion of course of prenatal corticosteroids and/or in utero transfer to perinatal center (European consensus Grade A)  second course of antenatal steroids may be appropriate if first course administered > 2-3 weeks earlier and infant < 33 weeks gestation when additional obstetric indication occurs (European consensus Grade A)  Reference - Neonatology 2013;103(4):353 PDF  Royal College of Obstetricians and Gynaecologists (RCOG) guideline on antenatal corticosteroids to reduce neonatal morbidity and mortality can be found at RCOG 2010 Oct PDF single antenatal steroid course  antenatal steroids for women at risk of preterm birth associated with reduced neonatal mortality and RDS (level 2 [mid-level] evidence)  based on Cochrane review of trials with methodologic limitations, and subsequent randomized trial  Cochrane review  systematic review of 21 randomized trials comparing antenatal corticosteroids (betamethasone, dexamethasone, hydrocortisone) vs. placebo or no treatment in 3,885 women (4,269 infants) with spontaneous preterm labor, preterm premature rupture of membranes (PPROM), or elective preterm delivery  methodologic limitations included unclear or inadequate allocation concealment in 13 trials, no placebo control in 8 trials, no intention-to-treat analysis in 11 trials; only 2 trials met all these quality criteria  search updated April 30, 2010 with 16 additional studies awaiting assessment  maternal outcomes comparing steroids vs. control (no significant differences)  maternal death in 0.53% vs. 0.56% (relative risk [RR] 0.98, 95% CI 0.06-15.5) in analysis of 3 trials with 365 women  chorioamnionitis in 7.4% vs. 8% (RR 0.91, 95% CI 0.7-1.18) in analysis of 12 trials with 2,485 women  puerperal sepsis in 11.5% vs. 8.7% (RR 1.35, 95% CI 0.93-1.95) in analysis of 8 trials with 1,003 women  for neonatal outcomes comparing steroids vs. control, steroids associated with reduced  combined fetal and neonatal death in analysis of 13 trials with 3,627 infants  RR 0.77 (95% CI 0.67-0.89)  NNT 16-49 assuming fetal and neonatal death in 18.8% of controls  neonatal death in analysis of 18 trials with 3,956 infants  RR 0.69 (95% CI 0.58-0.81)  NNT 16-36 assuming 14.8% neonatal death in controls  RDS in analysis of 21 trials with 4,038 infants, results limited by heterogeneity (p = 0.002)  RR 0.66 (95% CI 0.59-0.73)  NNT 4-15 assuming 26% RDS in controls  moderate-to-severe RDS in analysis of 6 trials with 1,686 infants, results may be limited by borderline statistical heterogeneity (p = 0.06)  RR 0.55 (95% CI 0.43-0.71)  NNT 11-21 assuming 17% moderate-to-severe RDS in controls  cerebroventricular hemorrhage in analysis of 13 trials with 2,872 infants  RR 0.54 (95% CI 0.43-0.69)  NNT 16-30 assuming 10.9% cerebroventricular hemorrhage in controls  severe cerebroventricular hemorrhage 4.5% vs. 16.8% (p < 0.0001, NNT 9) in analysis of 5 trials with 572 infants  RR 0.28 (95% CI 0.16-0.5)  NNT 9-12 assuming 16.8% severe cerebroventricular hemorrhage in controls  necrotizing enterocolitis in analysis of 8 trials with 1,675 infants  RR 0.46 (95% CI 0.29-0.74)  NNT 23-61 assuming 6.3% necrotizing enterocolitis in controls  systemic infection in first 48 hours in analysis of 5 trials with 1,319 infants  RR 0.56 (95% CI 0.38-0.85)  NNT 19-78 assuming 8.6% systemic infection in controls  steroids associated with nonsignificant reduction in chronic lung disease (CLD) in analysis of 6 trials with 818 infants, results limited by heterogeneity  no significant difference in mean birth weight in analysis of 11 trials with 3,586 infants  child outcomes comparing steroids vs. control (no significant differences)  death in childhood 3% vs. 4.2% (RR 0.68, 95% CI 0.36-1.27) in analysis of 4 trials with 1,010 children  neurodevelopmental delay 6% vs. 9.4% (RR 0.64, 95% CI 0.14-2.98) in 1 trial with 82 children  Reference - Cochrane Database Syst Rev 2006 Jul 19;(3):CD004454 (review updated 2010 Apr 30), also published in Obstet Gynecol 2007 Jan;109(1):189









in women at risk of preterm birth at < 24 weeks gestation, single course of antenatal corticosteroids may reduce neonatal mortality before hospital discharge (level 2 [mid-level] evidence)  based on systematic review of observational studies  systematic review of 17 observational studies comparing single course of antenatal corticosteroids within 7 days before birth vs. placebo or no treatment in women at risk of preterm birth at < 24 weeks gestation  all analyses included only neonates receiving active intensive treatment (resuscitation)  single course of antenatal corticosteroids associated with reduced risk of neonatal mortality before hospital discharge (adjusted odds ratio 0.48, 95% CI 0.38-0.61) in analysis of 4 studies with 3,610 infants  no significant differences in  respiratory distress syndrome in analysis of 2 studies with 861 infants  severe intraventricular hemorrhage in analysis of 2 studies with 859 infants  necrotizing enterocolitis in analysis of 3 studies with 1,051 infants  chronic lung disease in analysis of 3 studies with 1,184 infants  neurologic impairment at 18-22 months in 1 study with 601 infants  Reference - Obstet Gynecol 2016 Apr;127(4):715 single course of antenatal corticosteroids in women at high risk for preterm birth associated with improved neurodevelopmental outcomes in children born before 34 weeks gestation (level 2 [mid-level] evidence)  based on systematic review of mostly observational studies and with trial-specific quality measures not reported  systematic review of 14 randomized and nonrandomized clinical trials and observational studies evaluating neurodevelopment outcomes in children born before 34 weeks gestation whose mothers received a single course of betamethasone or dexamethasone (vs. placebo or no treatment) antenatally for threatened preterm birth  age range of offspring at follow-up 18 months to 31 years  single course of antenatal steroids associated with reduced risk of  cerebral palsy in analysis of 7 studies with 6,498 women  relative risk 0.678 (95% CI 0.564-0.815)  NNT 22-51 with cerebral palsy in 10.6% untreated children  psychomotor development index (PDI) < 70 in analysis of 2 studies with 4,018 women  relative risk 0.829 (95% CI 0.737-0.933)  NNT 14-56 with PDI < 70 in 26.6% untreated children  severe disability in analysis of 5 studies with 6,051 women  relative risk 0.787 (95% CI 0.729-0.850)  NNT 10-17 with severe disability in 39.2% untreated children  no significant differences in various scales of intelligence tests  Reference - Obstet Gynecol 2015 Jun;125(6):1385 single course of antenatal steroids not associated with adverse outcomes in long-term follow-up  based on 4 long-term follow-up studies  192 adult offspring (mean age 31 years) from 1,142 women participating in randomized trial of antenatal betamethasone vs. placebo  no significant differences in cognitive functioning, working memory and attention, psychiatric morbidity, handedness, or health-related quality of life  Reference - BMJ 2005 Sep 24;331(7518):665 full-text, editorial can be found in BMJ 2005 Sep 24;331(7518):645 full-text  81 adult offspring (aged 20-22 years) of 119 mothers at 26-32 weeks gestation (with threatened premature delivery) participating in randomized trial of antenatal steroids vs. placebo  no significant differences in medical or psychological variables  Reference - Pediatrics 2000 Jun;105(6):e77  cohort study of 130 surviving children (aged 14 years) of birth weight < 1,501 g (3.3 lbs), 53% of whom had antenatal corticosteroids  compared to offspring without antenatal corticosteroid exposure, steroid exposure had no obvious adverse effects on growth or on sensorineural, cognitive, or lung function  Reference - Pediatrics 2000 Jul;106(1):e2  534 offspring of mothers from randomized trial were exposed to antenatal betamethasone vs. placebo 2 doses 24 hours apart intramuscularly  no significant differences in cardiovascular risk factors in offspring aged 30 years  Reference - Lancet 2005 May 28;365(9474):1856 12-hour dosing interval for antenatal betamethasone appears as effective as 24-hour dosing interval for prevention of neonatal RDS, but may increase risk of necrotizing enterocolitis (level 2 [mid-level] evidence)  based on randomized trial without blinding





228 women pregnant with 260 singletons or twins at 23-34 weeks gestation and at risk for preterm delivery were randomized to receive 2 doses of betamethasone given 12 vs. 24 hours apart  comparing 12-hour dosing interval vs. 24-hour dosing interval  neonatal RDS in 36.5% vs. 37.3% (not significant)  necrotizing enterocolitis in 6.2% vs. 0% (p = 0.03, NNH 16)  Reference - Am J Obstet Gynecol 2012 Mar;206(3):201.e1 repeat antenatal steroid courses  repeat doses of prenatal corticosteroids may reduce RDS in infants at risk of preterm birth, but no significant differences in childhood outcomes at early childhood follow-up (18 months to 2 years corrected age) (level 2 [mid-level] evidence)  based on Cochrane review limited by clinical heterogeneity  systematic review of 10 randomized trials comparing repeat doses of corticosteroids vs. placebo or no treatment in 4,733 women and 5,700 infants who had already received corticosteroid at least 7 days previously and were still considered at risk of preterm birth  results limited by heterogeneity in prenatal corticosteroid regimens evaluated  most patients received betamethasone for repeat antenatal corticosteroid dosing, 14% in 1 trial received dexamethasone (6 mg intramuscularly every 12 hours up to 4 doses) due to unavailability of betamethasone  gestational age at enrollment typically was 25-33 weeks  betamethasone regimens included  betamethasone 12 mg/dose intramuscularly  for total of 2 doses 24 hours apart (2 trials, 1 trial summarized below)  for 2 doses 24 hours apart at weekly intervals until birth or 33-34 weeks gestation (5 trials)  for 2 doses 24 hours apart every 14 days until 33 weeks gestation or birth (1 trial)  betamethasone 12 mg intramuscularly for total of 1 dose (1 high-quality trial)  Celestone Chronodose 11.4 mg once weekly until delivery or 32 weeks gestation (1 high-quality trial, ACTORDS)  repeat doses of corticosteroids associated with  lower risk of RDS in analysis of 8 trials with 3,206 infants  risk ratio 0.83 (95% CI 0.75-0.91)  NNT 17 (95% CI 11-32)  results in 2 high-quality trials suggest outcomes dependent on number of repeat corticosteroid injections  repeat doses of corticosteroids as multicourse regimen significantly decreased RDS in 1 trial with 1,144 women  no significant difference with single repeat dose of betamethasone 12 mg intramuscularly in 1 trial with 326 women  decreased mean birth weight in analysis of 9 trials with 5,626 infants  mean difference -76 g (-0.2 lbs) (95% CI -118 g to -34 g [-0.3 lbs to -.07 lbs])  when birth weight adjusted for gestational age, no significant difference found between groups in analysis of 2 trials with 1,256 infants  nonsignificantly lower rate of vaginal birth (RR 0.93, 95% CI 0.87-1) and nonsignificantly higher rate of cesarean section (RR 1.05, 95% CI 0.99-1.1) in analyses of 7 trials with 4,062 women  no significant differences in  perinatal or early postpartum outcomes including  severe lung disease in analysis of 6 trials with 4,826 infants, results limited by significant heterogeneity  fetal or neonatal mortality in analysis of 9 trials with 5,554 infants  chronic lung disease in analysis of 8 trials with 5,393 infants  intraventricular hemorrhage in analysis of 6 trials with 3,065 infants  chorioamnionitis in analysis of 6 trials with 4,261 women  puerperal sepsis in analysis of 5 trials with 3,091 women  early childhood outcomes (18 months to 2 years corrected age) including  total mortality in analysis of 4 trials with 4,370 children  survival free of any disability in analysis of 2 trials with 3,155 children  survival free of major neurosensory disability in analysis of 2 trials with 1,317 children, results limited by significant heterogeneity  Reference - Cochrane Database Syst Rev 2015 Jul 5;(7):CD003935  DynaMed commentary -- 3 trials used single steroid course regimens which may have diminished effects in repeated dose group







repeat doses of antenatal corticosteroids in women at risk for very preterm birth do not appear to affect neurosensory, cognitive, or behavioral outcomes in children surviving to 6-8 years corrected age (level 2 [midlevel] evidence)  based on post hoc follow-up analysis of ACTORDS trial  982 pregnant women at risk of very preterm birth who had corticosteroid injection ≥ 1 week before were randomized to betamethasone (Celestone Chronodose 11.4 mg) intramuscularly vs. saline placebo with repeated injection weekly if risk of preterm birth persisted  of 1,144 live births, 963 children who survived to 6-8 years corrected age were assessed  neurosensory disability-free survival in 78% with repeat betamethasone vs. 77% with placebo (not significant)  no significant difference in intellectual impairment, cognitive function, behavior, or general health including lung function and blood pressure  Reference - Pediatrics 2016 Oct;138(4):e20160947 "rescue course" of antenatal corticosteroids associated with reduced RDS and ventilator and surfactant use (level 2 [mid-level] evidence)  based on randomized trial with allocation concealment not stated  437 women with singletons or twins < 33 weeks gestation with recurring threat of preterm delivery in upcoming week were randomized to single rescue course of betamethasone vs. placebo intramuscularly  all women had completed single course of antenatal corticosteroids before 30 weeks gestation and ≤ 14 days before inclusion  betamethasone given as two 12 mg doses 24 hours apart ("rescue dose")  delivery at < 34 weeks gestation in 55% in each group  known outcomes in 97% of 577 delivered infants  comparing rescue steroid group vs. placebo  RDS (in infants born at < 34 weeks gestation) in 41.4% vs. 61.6% (p = 0.002, NNT 5)  RDS (in all infants) in 30.2% vs. 41.3% (p = 0.026, NNT 9)  surfactant use (in infants born at < 34 weeks gestation) in 37.7% vs. 55.5% (p = 0.004, NNT 6)  surfactant use (in all infants) in 25.6% vs. 35.4% (p = 0.038, NNT 11)  ventilator support (in infants born at < 34 weeks gestation) in 37.6% vs. 52.9% (p = 0.023, NNT 7)  ventilator support (in all infants) in 26.2% vs. 34.8% (p = 0.088)  no significant differences (in both infants born only at < 34 weeks gestation or in all infants) in  bronchopulmonary dysplasia (BPD)  severe intraventricular hemorrhage  periventricular leukomalacia  blood culture-proven sepsis  necrotizing enterocolitis  perinatal death (stillbirth or death before neonatal hospital discharge)  Reference - Am J Obstet Gynecol 2009 Mar;200(3):248.e1, editorial can be found in Am J Obstet Gynecol 2009 Mar;200(3):217 multiple courses of antenatal corticosteroids may not further reduce mortality or neurodevelopmental disability compared to single course in children aged 5 years (level 2 [mid-level] evidence)  based on follow-up of randomized trial  1,858 pregnant women (2,318 fetuses) between 25 and 32 weeks gestation at risk of preterm birth were randomized to antenatal betamethasone 12 mg intramuscularly 2 doses 1 day apart (multiple course) vs. betamethasone 12 mg intramuscularly once (single course)  multiple course administered every 2 weeks until 33 weeks gestation or birth  single course group had placebo injections on same days as multiple course group after initial betamethasone dose  in original trial  no significant differences in perinatal or neonatal mortality and morbidity (severe respiratory distress syndrome, intraventricular hemorrhage, periventricular leukomalacia, bronchopulmonary dysplasia, or necrotizing enterocolitis)  significant decrease in weight, length, and head circumference at birth comparing multiple-course vs. singlecourse betamethasone  1,724 women (2,141 eligible children) completed follow-up at age 5 years  data from 1,719 children contributed to primary outcome (death or neurodevelopmental disability, defined as deficit in neuromotor, neurosensory, or neurocognitive/neurobehavioral function)  at 5-year follow-up  primary outcome in 24.9% with multiple-course vs. 24.8% with single-course betamethasone (not significant)  no significant difference in weight, height, or head circumference  Reference - MACS-5 trial (JAMA Pediatr 2013 Dec;167(12):1102), commentary can be found in JAMA Pediatr 2014 Apr;168(4):389





multiple courses of antenatal corticosteroids associated with earlier birth and dose-dependent decrease in fetal growth in women at high risk of preterm delivery (level 2 [mid-level] evidence)  based on secondary analysis of randomized trial  1,858 women at 25-33 weeks gestation who had not delivered 14-21 days after single course of antenatal corticosteroids and remained at high risk of preterm delivery randomized to additional courses of antenatal corticosteroids vs. placebo every 14 days until week 33 or delivery  additional antenatal corticosteroid course consisted of 2 doses betamethasone 12 mg intramuscularly given 24 hours apart  mean gestational age at birth 34.5 weeks for additional courses antenatal corticosteroids vs. 34.9 weeks for placebo (p < 0.001)  after controlling for gestational age at birth and other confounding factors, each additional course of antenatal corticosteroids associated with nonsignificant incremental decrease in birth weight, length, and head circumference  Reference - Obstet Gynecol 2012 May;119(5):917 comparative efficacy and safety among prenatal steroids  antenatal dexamethasone associated with reduced risk of intraventricular hemorrhage compared to antenatal betamethasone (level 2 [mid-level] evidence)  based on Cochrane review of trials with methodologic limitations  systematic review of 12 randomized or quasi-randomized trials evaluating antenatal corticosteroids in 1,557 women (1,661 infants) at risk of preterm birth  all trials had ≥ 1 limitation including  unclear or inadequate allocation concealment  lack of or unclear blinding  high loss to follow-up  comparing dexamethasone to betamethasone  dexamethasone associated with decreased intraventricular hemorrhage overall in analysis of 4 trials with 549 infants  risk ratio 0.44 (95% CI 0.21-0.92)  NNT 18-179 with intraventricular hemorrhage in 7% of betamethasone group  no significant differences in neonatal death, respiratory distress syndrome, bronchopulmonary dysplasia, severe intraventricular hemorrhage, periventricular leukomalacia, and mean birth weight  no significant difference in neonatal intensive care unit (NICU) admission in 1 trial with 240 neonates, but dexamethasone group had increased NICU admission in 1 trial with 105 neonates  for outcome of length of admission to birth in 1 trial with 240 women  antenatal dexamethasone group had significantly longer length of admission to birth in subgroup with intact membranes  no significant difference in subgroup with ruptured membranes  Reference - Cochrane Database Syst Rev 2013 Aug 29;(8):CD006764  dexamethasone may reduce intraventricular hemorrhage compared to betamethasone (level 2 [mid-level] evidence)  based on randomized trial without clear accounting for all neonates for some primary outcomes (largest trial in Cochrane review)  299 women at risk of preterm delivery randomized to dexamethasone (6 mg at 0, 12, 24, and 36 hours) vs. betamethasone (12 mg at 0 and 24 hours) intramuscularly  less than 70% had results for intraventricular hemorrhage  patients treated with betamethasone were given placebo injection at 12 and 36 hours to maintain blinding  comparing all 178 infants exposed to dexamethasone vs. all 181 infants exposed to betamethasone  no significant differences in neonatal death, RDS, BPD, necrotizing enterocolitis, retinopathy of prematurity, patent ductus arteriosus (PDA), neonatal sepsis  comparing outcomes selectively reported in 105 infants exposed to dexamethasone vs. 100 infants exposed to betamethasone  5.7% vs. 17% had intraventricular hemorrhage of any grade (p = 0.02, NNT 9)  1.9% vs. 7% had grade 3 or 4 intraventricular hemorrhage (p = 0.09)  6.7% vs. 18% had any brain lesion (p = 0.02, NNT 9)  1.9% vs. 4% had periventricular leukomalacia (not significant)  Reference - Betacode trial (Obstet Gynecol 2007 Jul;110(1):26), editorial can be found in Obstet Gynecol 2007 Jul;110(1):7, commentary can be found in Obstet Gynecol 2007 Oct;110(4):930  prenatal dexamethasone exposure may be associated with higher risk for neurodevelopmental abnormalities than prenatal betamethasone exposure (level 2 [mid-level] evidence)  based on 2 cohort studies



prenatal exposure to dexamethasone in extremely low-birth-weight infants may be associated with worse neurodevelopmental outcomes at 18-22 months than prenatal betamethasone (level 2 [mid-level] evidence)  based on cohort study  1,124 extremely low-birth-weight infants born 2002-2003 had neurodevelopmental assessment at 18-22 months corrected age No SteroidDexamethasone Betamethasone Cerebral palsy 13.1% 13.5% 9.9% (not significant) PDI < 70 20% 24% 19% (not significant) PDI ≥ 85 56% 51% 64% (p < 0.05)* Deafness 3.3% 3.2% 0.9% (p < 0.05)* Neurodevelopmental impairment45% 41% 34% (p < 0.05)* Unimpaired 21% 26% 40% (p < 0.05)* Abbreviation: PDI, Psychomotor Development Index.

* p values for dexamethasone vs. betamethasone comparisons. Neurodevelopmental Assessments by Prenatal Steroid Exposure:  Reference - Pediatrics 2008 Feb;121(2):289 full-text  multiple antenatal courses of dexamethasone (but not betamethasone) associated with increased risk of leukomalacia and neurodevelopmental abnormalities at 2 years (level 2 [mid-level] evidence)  based on prospective cohort study  201 preterm singleton infants who received ≥ 1 course of antenatal steroids at 24-34 weeks gestation were evaluated  138 infants (68.7%) received betamethasone  63 infants (31.3%) received dexamethasone  overall prevalence of infant leukomalacia  25.9% after 1 complete course of corticosteroids  40% after 1 additional course  42.3% after 2 courses  44.4% after > 2 additional courses  overall prevalence of neurodevelopmental abnormalities  18% after 1 complete course of corticosteroids  21.4% after 1 additional course  29.2% after 2 courses  34.8% after > 2 additional courses  comparing dexamethasone vs. betamethasone among 70 patients with multiple corticosteroid courses  leukomalacia in 58% vs. 30% (p < 0.05)  neurodevelopmental abnormalities in 43% vs. 17% (p < 0.05)  Reference - Am J Obstet Gynecol 2004 Jul;191(1):217  no randomized trials found comparing maternal to direct fetal routes of administration of antenatal corticosteroids  based on Cochrane review  Reference - Cochrane Database Syst Rev 2011 Sep 7;(9):CD008981 Corticosteroids after 34 weeks gestation  recommendations for antenatal steroids in pregnancies between 34 and 36 5/7 weeks gestation  Society for Maternal and Fetal Medicine statement on use of antenatal corticosteroids in the late preterm birth period in women at risk for preterm delivery  betamethasone is recommended for singleton pregnancies between 34 weeks and 36 completed weeks of gestation at high risk for preterm birth within next 7 days but before 37 weeks gestation  2 doses of 12 mg intramuscularly given 24 hours apart is recommended dosing regimen  Reference - Am J Obstet Gynecol 2016 Aug;215(2):B13  American College of Obstetricians and Gynecologists (ACOG) practice advisory for antenatal corticosteroid therapy for fetal maturation in late preterm birth  consider administration of betamethasone in women with a singleton pregnancy between 34 0/7 and 36 6/7 weeks gestation at imminent risk of preterm birth within 7 days  administration of betamethasone in the late preterm period should not be used if pregnancy was already exposed to antenatal corticosteroids  late term corticosteroids not indicated in women with chorioamnionitis  Reference - American College of Obstetricians and Gynecologists (ACOG) practice advisory 2016 Apr 4  antenatal corticosteroids given at ≥ 34 weeks gestation may decrease risk of transient tachypnea and respiratory distress syndrome in singleton neonates (level 2 [mid-level] evidence)

 



based on systematic review limited by heterogeneity systematic review of 6 randomized trials comparing antenatal corticosteroids vs. placebo or no treatment in 5,698 women with singleton pregnancy at ≥ 34 weeks gestation  3 trials enrolled women having planned cesarean delivery at ≥ 37 weeks and 3 trials enrolled women at 34-36 weeks gestation at risk of imminent late preterm delivery  corticosteroids included betamethasone (4 trials) and dexamethasone (2 trials)  antenatal corticosteroids given at ≥ 34 weeks gestation associated with  decreased risk of transient tachypnea in analysis of 5 trials with 5,598 patients, results limited by significant heterogeneity  risk ratio (RR) 0.56 (95% CI 0.37-0.86)  NNT 18-80 with transient tachypnea in 9% of placebo or no treatment group  decreased risk of respiratory distress syndrome in analysis of all trials  RR 0.74 (95% CI 0.61-0.91)  NNT 37-159 with respiratory distress syndrome in 7% of placebo or no treatment group  decreased mechanical ventilation use in analysis of 4 trials with 5,146 patients, results limited by significant heterogeneity  RR 0.52 (95% CI 0.36-0.76)  NNT 49-131 with mechanical ventilation in 3.2% of placebo or no treatment group  no significant differences in  neonatal mortality in analysis of all trials  admission to neonatal intensive care unit in analysis of all trials, results limited by significant heterogeneity  mask ventilation use in analysis of 4 trials with 4,019 patients, results limited by significant heterogeneity  Reference - BMJ 2016 Oct 12;355:i5044 full-text antenatal betamethasone improves respiratory outcomes in singleton late preterm infants (level 1 [likely reliable] evidence)  based on randomized trial  2,831 women with singleton pregnancy between 34 and 36 5/7 weeks gestation at high risk of late preterm delivery were randomized to betamethasone (2 doses of 12 mg intramuscularly given 24 hours apart) vs. placebo  high risk of late preterm delivery (up to 36 6/7 weeks gestation) defined as preterm labor with intact membranes plus ≥ 3 cm dilation or 75% cervical effacement, spontaneous rupture of membranes, or expected induction or cesarean section for any indication 24 hours to 7 days after randomization  women were excluded for prior antenatal glucocorticoid use, expected delivery in < 12 hours, or lack of gestational-dating results before 32 weeks in women with known date of last menstrual period or 24 weeks for women with unknown date of last menstrual period  59.6% received both doses of trial medication  94.6% of 1,145 women not receiving second dose delivered within 24 hours of first dose  99.9% of infants were included in analysis  primary outcome was composite of need for respiratory support (use of continuous positive airway pressure [CPAP] or high-flow nasal cannula for ≥ 2 continuous hours, supplemental oxygen with fraction of inspired oxygen ≥ 30% for ≥ 4 continuous hours, extracorporeal membrane oxygenation [ECMO], or mechanical ventilation), neonatal death within 72 hours of delivery, or stillbirth  severe respiratory complications defined as CPAP or high-flow nasal cannula for ≥ 12 continuous hours, supplemental oxygen with fraction of inspired oxygen ≥ 0.3 for ≥ 24 continuous hours, ECMO, mechanical ventilation, stillbirth, or neonatal death within 72 hours of delivery  comparing betamethasone vs. placebo  primary outcome in 11.6% vs. 14.4% (p = 0.02, NNT 36)  CPAP or high-flow nasal cannula for  ≥ 2 continuous hours in 10.2% vs. 13.1% (p = 0.01, NNT 35)  ≥ 12 continuous hours in 6.5% vs. 10.5% (p < 0.001, NNT 25)  severe respiratory complications in 8.1% vs. 12.1% (p < 0.001, NNT 25)  need for resuscitation at birth in 14.5% vs. 18.7% (p = 0.003, NNT 24)  transient tachypnea of newborn in 6.7% vs. 9.9% (p = 0.002, NNT 32)  surfactant use in 1.8% vs. 3.1% (p = 0.03)  bronchopulmonary dysplasia in 0.1% vs. 0.6% (p = 0.04)  neonatal hypoglycemia in 24% vs. 15% (p < 0.001, NNH 12)  no stillbirths, neonatal deaths within 72 hours, or need for ECMO in either group  no significant differences in supplemental oxygen with fraction of inspired oxygen ≥ 30% for ≥ 4 hours or ≥ 24 hours, mechanical ventilation, respiratory distress syndrome, apnea, pneumonia, pulmonary air leak, neonatal sepsis, or maternal outcomes

 

95% of maternal adverse events were local injection site reactions Reference - ALPS Trial ( N Engl J Med 2016 Apr 7;374(14):1311) Multifaceted intervention including corticosteroids  multifaceted intervention for health providers to promote use of single course of antenatal corticosteroids may not reduce neonatal mortality due to preterm birth in low- and middle-income countries (level 2 [mid-level] evidence)  based on cluster-randomized trial with unclear blinding of outcome assessors  102 rural and semiurban health facilities in low- and middle-income countries were randomized to multifaceted intervention vs. control for 18 months  multifaceted intervention consisted of health provider training to improve identification of women at < 36 weeks gestation at risk of preterm birth and facilitate appropriate use of antenatal corticosteroids (single 4-dose course of dexamethasone 6 mg every 12 hours)  all clusters received training in essential newborn care only  99,742 women with 98,137 live births were included in analyses  birth weight < 5th percentile used as surrogate for preterm birth (4.5% of live births)  comparing multifaceted intervention vs. usual care  28-day neonatal mortality per 1,000 live births in preterm neonates 225 vs. 232 (not significant)  overall 28-day neonatal mortality per 1,000 live births 27.4 vs. 23.9 (p = 0.0127)  antenatal corticosteroid use in women delivering preterm neonates 45% vs. 10% (p < 0.0001)  suspected maternal infection in women delivering preterm neonates in 10% vs. 6% (p < 0.0001, NNH 25)  overall suspected maternal infection in 3% vs. 2% (p < 0.0001, NNH 100)  Reference - ACT trial (Lancet 2015 Feb 14;385(9968):629), editorial can be found in Lancet 2015 Feb 14;385(9968):585 Magnesium  if birth anticipated < 32 weeks, administer magnesium sulfate to reduce severity and risk of cerebral palsy in neonate (ACOG Level A)(1)  World Health Organization recommends magnesium sulfate in women at risk of imminent preterm birth < 32 weeks gestational age for prevention of cerebral palsy (WHO Strong recommendation, Moderate-quality evidence) (WHO 2015 PDF)  recommended dose 4-6 g IV bolus for 20 minutes followed by 1-2 g/hour (max 3 g/hour)(2)  periviable birth  magnesium sulfate for neuroprotection for threatened and imminent periviable birth based on best estimate of gestational age(4)  recommended for pregnancies ≥ 24 weeks gestation ( SMFM/ACOG Grade 1B)  consider for pregnancies 23 weeks gestation ( SMFM/ACOG Grade 2B)  not recommended for pregnancies > 20 weeks gestation but < 23 weeks gestation (SMFM/ACOG Grade 1A)  antenatal magnesium sulfate in women at risk for preterm birth decreases risk of cerebral palsy and substantial gross motor dysfunction (level 1 [likely reliable] evidence)  based on Cochrane review  systematic review of 5 randomized trials evaluating antenatal magnesium sulfate for fetal neuroprotection in women at risk of imminent delivery, including 6,145 infants  trials excluded if magnesium sulfate was used primarily for tocolysis, prevention or treatment of eclampsia, or maintenance therapy after preterm labor  4 trials specifically targeted women likely to have preterm birth and magnesium was being used for neuroprotection  outcome measures varied across trials  most trials used magnesium sulfate 4 g IV bolus followed by maintenance infusion of 1-3 g/hour for 12-24 hours or until birth  magnesium sulfate associated with reduced risk of  cerebral palsy in analysis of 5 trials with 6,145 infants  relative risk (RR) 0.68 (95% CI 0.54-0.87)  NNT 43-154 with cerebral palsy in 5% of controls  severe gross motor dysfunction (defined as child not walking at ≥ 2 years or unable to grasp/release small block with both hands) in analysis of 4 trials with 5,980 infants  RR 0.61 (95% CI 0.44-0.85)  NNT 60-222 with severe gross motor dysfunction in 3.1% of controls  no significant differences in major maternal complications or pediatric mortality and morbidity  Reference - Cochrane Database Syst Rev 2009 Jan 21;(1):CD004661, commentary can be found in Evid Based Med 2009 Oct;14(5):141  no completed randomized trials found comparing different magnesium sulfate regimens for neuroprotection of fetus in women at risk of preterm birth  based on Cochrane review  Reference - Cochrane Database Syst Rev 2012 Feb 15;(2):CD009302

Thyrotropin-releasing hormones addition of thyrotropin-releasing hormones to steroids in women at risk of preterm delivery does not reduce mortality or adverse fetal outcomes, and causes maternal and neonatal adverse effects (level 1 [likely reliable] evidence)  based on Cochrane review  systematic review of 15 randomized trials evaluating addition of thyrotropin-releasing hormones (TRH) to prenatal corticosteroids in > 4,600 women at risk of preterm birth  TRH dosing regimens varied, majority of trials used 400 mcg every 8 hours (up to 4-6 doses)  addition of TRH to steroids did not significantly reduce  mortality in analysis of 6 trials with 3,694 infants  respiratory distress syndrome in analysis of 9 trials with 3,833 infants  chronic oxygen dependence in analysis of 5 trials with 2,511 infants  TRH associated with increased risk of  maternal adverse effects (including nausea, vomiting, lightheadedness, facial flushing, and urgency of micturition)  infants needing ventilation (risk ratio [RR] 1.16, 95% CI 1.03-1.29) in analysis of 3 trials with 1,969 infants  low Apgar score at 5 minutes (RR 1.48, 95% CI 1.14-1.92) in analysis of 3 trials with 1,969 infants  adverse infant neurological outcomes at 12-month follow-up in 3 trials  Reference - Cochrane Database Syst Rev 2013 Nov 21;(11):CD000019 Phenobarbital  prophylactic maternal phenobarbital prior to very preterm birth may not prevent periventricular hemorrhage in infants (level 2 [mid-level] evidence)  based on Cochrane review of trials with methodologic limitations and inconsistent results between analysis of all trials and analysis of only higher-quality trials  systematic review of 9 randomized trials evaluating phenobarbital administration in 1,752 women at risk of delivering < 34 weeks gestation  most trials had ≥ 1 of the following limitations  unclear or inadequate allocation concealment  high loss to follow-up  2 high-quality trials both had unclear randomization methods  outcome of periventricular hemorrhage includes intracranial hemorrhage assessed in 2 studies, intraventricular hemorrhage assessed in 4 studies, and periventricular hemorrhage (PVH) assessed in 3 studies  comparing phenobarbital vs. control  no significant differences in any PVH or severe PVH in analysis limited to 2 higher quality trials with 945 women  phenobarbital associated with  reduced PVH in analysis of 9 trials with 1,591 women  risk ratio (RR) 0.65 (95% CI 0.5-0.83)  NNT 6-17 with PVH in 36% of controls  results limited by significant heterogeneity  reduced severe PVH (grades III-IV) in analysis of 8 trials with 1,527 women  RR 0.41 (95% CI 0.2-0.85)  NNT 14-74 with severe PVH in 9% of controls  results limited by significant heterogeneity  no significant difference in neurodevelopmental abnormalities at 18-24 months or 7-year follow-up  Reference - Cochrane Database Syst Rev 2011 Mar 16;(1):CD000164 Vitamin K  effect of prophylactic maternal vitamin K prior to very preterm birth for preventing severe periventricular hemorrhage in infants is uncertain (level 2 [mid-level] evidence)  based on Cochrane review of trials with methodologic limitations and inconsistent results between overall analysis and analysis excluding quasi-randomized trials  systematic review of 5 randomized and 2 quasi-randomized trials evaluating vitamin K administered parenterally or orally vs. placebo or no treatment in 851 women at risk of imminent preterm birth  unclear or inadequate allocation concealment in most trials  compared to controls, vitamin K associated with  reduced severe periventricular hemorrhage (PVH) (grades III-IV) in analysis of all trials (randomized and quasirandomized) with 851 infants, results limited by significant heterogeneity  risk ratio (RR) 0.58 (95% CI 0.37-0.91)  NNT 14-93 with severe PVH in 12% of control group  no significant difference in severe PVH in analysis of 5 randomized trials (RR 0.87, 95% CI 0.6-1.26)





lower Bayley Mental Developmental Index at 2 years (p = 0.021) in 1 trial with 121 children and high loss to followup  Reference - Cochrane Database Syst Rev 2010 Jan 20;(1):CD000229 Surgery and procedures  no evidence to suggest that prophylactic episiotomy, forceps delivery, or cesarean section (other than for nonvertex position) improve neonatal outcomes in women with preterm delivery(2)  World Health Organization does not recommend routine delivery by caesarean section to improve preterm neonatal outcomes, regardless of cephalic or breech presentation (WHO Conditional recommendation, Very low-quality evidence) (WHO 2015 PDF)  cesarean delivery for fetal indication (such as abnormal heart rate or biophysical testing) in patients with threatened and imminent periviable birth based on best estimate of gestational age(4)  recommended for pregnancies ≥ 25 weeks gestation (SMFM/ACOG Grade 1B)  consider for pregnancies 23 weeks gestation (SMFM/ACOG Grade 1B) and 24 weeks gestation (SMFM/ACOG Grade 1B)  not recommended for pregnancies > 20 weeks gestation but < 23 weeks gestation (SMFM/ACOG Grade 1A) policy of planned immediate cesarean section associated with nonsignificant decrease in risk of perinatal mortality compared to policy of planned vaginal birth in women with imminent preterm delivery (level 2 [mid-level] evidence)  based on Cochrane review of trials with early termination  systematic review of 6 randomized trials comparing policy of planned immediate cesarean section vs. policy of planned vaginal birth in 122 women with preterm labor  data from only 4 trials could be used in analyses, all trials terminated early due to recruitment problems  planned immediate cesarean section associated with  nonsignificant decrease in perinatal mortality (risk ratio [RR] 0.29, 95% CI 0.07-1.14) in analysis of 3 trials with 89 infants  increase in major postpartum maternal complications (RR 7.21, 95% CI 1.37-38.08, with major postpartum maternal complications in 0% of planned vaginal birth group) in analysis of 4 trials with 116 women  no significant differences in  birth injury to infant in 1 trial with 38 women  birth asphyxia in 1 trial with 12 women  Reference - Cochrane Database Syst Rev 2013 Sep 12;(9):CD000078 Other management  historical treatments no longer recommended in women with preterm labor due to limited evidence for efficacy include(1)  routine bed rest (ACOG Level B)  sexual abstention  relaxation therapy may not reduce preterm birth in women in preterm labor (level 2 [mid-level] evidence)  based on Cochrane review of trials with inadequate or unclear allocation concealment  systematic review of 11 randomized trials evaluating relaxation therapy for prevention or treatment of preterm labor in 833 pregnant women  4 trials evaluated relaxation therapy in women in preterm labor  addition of music therapy to standard treatment did not significantly reduce preterm birth in 1 trial with 120 women  comparing relaxation therapy to no treatment, no significant differences in  pregnancy prolongation or gestational age at birth in 1 trial with 107 women  birth weight in analysis of 2 trials with 121 infants  Reference - Cochrane Database Syst Rev 2012 Aug 15;(8):CD007426 Complications and Prognosis Complications  prematurity(2, 3)  low birth weight (< 2,500 g)(2)  admission to neonatal intensive care unit (ICU)(2)  see Evaluation and management of the premature infant for specific complications of premature delivery  periviable cesarean delivery associated with increased likelihood of vertical uterine incision extending into upper muscular part of uterus, resulting in increased risk of perioperative morbidity and uterine rupture in subsequent pregnancy(4) Prognosis  general prognosis  < 10% of women diagnosed with preterm labor reported to give birth within 7 days(1)  about 30% of preterm labors reported to spontaneously resolve without treatment(1)  about 50% of women hospitalized for preterm labor reported to give birth at term(1)  most women with singleton pregnancy and threatened preterm labor appear unlikely to deliver within 1 week  based on systematic review of prognostic cohort and case-control studies





systematic review of 28 cohort and case-control studies evaluating transvaginal ultrasound assessment of cervical length for prediction of preterm birth in women with singleton pregnancy, intact membrane, and preterm labor symptoms  pooled prevalence 11.1% for birth within 1 week from presentation  rate of birth within 48 hours of presentation 7.1% in analysis of 3 studies in 1,266 women  rate of birth within 7 days of presentation 11% in analysis of 6 studies in 1,781 women  prevalence of birth within 7 days varied greatly among studies, ranging from 8.4% to 31%  Reference - Ultrasound Obstet Gynecol 2010 Jan;35(1):54 full-text  cervical length on ultrasound appears significantly associated with risk of preterm delivery, but predictive accuracy appears low in women with threatened preterm labor (level 2 [mid-level] evidence)  based on retrospective prognostic cohort study  1,077 pregnant women with preterm labor at < 34 weeks gestation had ultrasound determination of cervical length  correlation between cervical length and time to delivery was significant but weak (each additional 2 mm associated with 1 day increase in interval) (p < 0.001)  Reference - Obstet Gynecol 2013 Dec;122(6):1279 prognostic models  prognostic model for threatened preterm delivery predicts preterm delivery within 48 hours (level 1 [likely reliable] evidence) and may predict preterm delivery before 32 weeks gestation (level 2 [mid-level] evidence)  based on derivation and validation cohort study  derivation cohort included 737 women (mean age 29 years) at 22-32 weeks gestation who were transferred from maternity care center to tertiary care center for threatened preterm delivery  157 (21.3%) women delivered within 48 hours after transfer  317 (43%) delivered at < 32 weeks gestation  validation cohort included 169 similar women at a different tertiary care center  40 (23.7%) women delivered within 48 hours after transfer  59 (34.9%) delivered at < 32 weeks gestation  prognostic model risk calculator includes  gestational age  sonographic cervical length  vaginal bleeding  preterm premature rupture of membranes (PPROM)  uterine contractions  number of fetus  in validation cohort, risk score had  good correlation between predicted and observed risk of delivery within 48 hours of transfer  good overall correlation between predicted and observed risk of delivery at < 32 weeks gestation, but may overestimate risk in low-risk women and underestimate risk in high-risk women  online risk calculator can be found at Probability of Preterm Delivery  Reference - Am J Obstet Gynecol 2011 Mar;204(3):242.e1  prognostic models may determine likelihood of preterm delivery within 10 days of labor symptoms or before 37 weeks gestation (level 2 [mid-level] evidence)  based on prognostic cohort study without independent validation  583 women with singleton pregnancies at 24-34 weeks gestation with symptoms of preterm labor (contractions, cramping, asymptomatic vaginal bleeding, vaginal pressure, and abdominal or back pain) were analyzed  15.4% delivered within 10 days of symptom onset  35% delivered < 37 weeks gestation  risk factors significantly associated with preterm delivery and points assigned to derive risk scores  for delivery within 10 days of symptom onset (total score 0-16 points)  no prenatal care - 4 points  initial cervical dilation - 2 points if 2 cm to < 3 cm, 4 points if 3 cm to < 4 cm, 10 points if ≥ 4 cm  tobacco use - 2 points  for delivery < 37 weeks gestation (total score 0-19 points)  initial cervical dilation - 2 points if 2 cm to < 3 cm, 4 points if 3 cm to < 4 cm, 15 points if ≥ 4 cm  obstetric history - 1 point if previous full-term birth or previous full-term birth and previous preterm birth, 2 points if previous preterm birth  tobacco use - 2 points  accuracy of risk scores for determining likelihood of preterm delivery

Test Characteristics Delivery Within 10 Days of Symptom Onset Delivery < 37 Weeks Gestation Cutoff point ≥ 10 points ≥ 8 points Sensitivity 32% 22% Specificity 95% 97% Positive predictive value 53% 80% Negative predictive value 89% 70% Correctly classified patients85% 71% Results:  Reference - Obstet Gynecol 2012 Jun;119(6):1119  prognostic model using serum proteins and cervical length may help determine likelihood of spontaneous preterm delivery within 7 days in women with threatened preterm labor (level 2 [mid-level] evidence)  based on prognostic cohort study without independent validation  142 women with singleton pregnancies at 22-33 weeks gestational age with symptoms of preterm labor were assessed for 27 serum proteins and cervical length  40% delivered within 7 days  prognostic models based on serum protein levels and cervical length were developed and assessed for predicting preterm delivery  most accurate model for determining likelihood of preterm delivery was combined presence of  serum interleukin-10 ≥ 48 pg/mL  regulated on activation normal T-expressed and secreted (RANTES) ≥ 49,293 pg/mL  cervical length ≤ 18 mm  performance of prognostic model for determining likelihood of spontaneous preterm delivery within 7 days in women with threatened preterm labor  sensitivity 73.8%  specificity 87.3%  positive predictive value 75.6%  negative predictive value 86.2%  positive likelihood ratio 5.83  negative likelihood ratio 0.3  Reference - BJOG 2012 Jun;119(7):866, commentary can be found in BJOG 2012 Nov;119(12):1544  see Evaluation and management of the premature infant for neonatal outcomes Prevention and Screening Prevention  progesterone  injectable or vaginal progesterone appears to reduce risk of preterm birth (level 2 [mid-level] evidence)  progesterone (injectable or vaginal) does not prevent premature birth in twin pregnancies (level 1 [likely reliable] evidence)  cervical cerclage  for women with history of preterm delivery or spontaneous second trimester loss  offer cerclage to women with any of  ≥ 1 second trimester pregnancy losses related to painless cervical dilation and in absence of labor or abruptio placentae (ACOG Level B)  prior cerclage due to painless cervical dilation in second trimester (ACOG Level B)  ≥ 3 previous preterm births and/or second trimester losses (RCOG Grade B)  physical exam suggests current painless cervical dilation in second trimester (ACOG Level B)  ultrasound findings suggesting short (< 25 mm) cervical length before 24 weeks gestation if (ACOG Level A; RCOG Grade A )  prior spontaneous preterm birth < 34 weeks gestation and  current pregnancy is singleton  do not routinely offer cerclage to women with ≤ 2 previous preterm births and/or second trimester losses (RCOG Grade B)  cerclage may reduce preterm delivery with conflicting evidence for reducing perinatal morbidity and mortality in women with high risk for preterm birth (level 2 [mid-level] evidence)  cerclage not recommended for  women without history of spontaneous preterm delivery or second trimester loss with incidental finding of cervical length ≤ 25 mm (ACOG Level B; RCOG Grade B)  women with funneling of cervix (dilation of internal os) without cervix ≤ 25 mm (RCOG Grade C)  women with previous cervical surgery (cone biopsy, large loop excision of the transformation zone, laser ablation, diathermy) (RCOG Grade B)  sole history of loop electrosurgical excision procedure, cone biopsy, or mullerian anomaly (ACOG Level B)

  

 

women with multiple pregnancies (RCOG Grade B) twin pregnancy and short cervical length (< 25 mm) detected on ultrasound (ACOG Level B) cerclage may increase preterm delivery in women with multiple gestation (level 2 [mid-level] evidence)  contraindications to cerclage include  active preterm labor  evidence of chorioamnionitis  persistent vaginal bleeding  preterm premature rupture of membranes addition of cervical pessary to expectant management reduces risk of preterm birth in women with cervical length ≤ 25 mm (level 1 [likely reliable] evidence) antibiotics

  



antibiotics for bacterial vaginosis given before 20 weeks gestation have conflicting evidence for reduction of preterm birth, but may reduce risk of late miscarriage (level 2 [mid-level] evidence) antibiotic treatment of asymptomatic bacteriuria does not appear to reduce preterm delivery rates (level 2 [mid-level] evidence) interpregnancy antibiotics do not appear to prevent preterm birth or miscarriage in women with previous preterm birth (level 2 [mid-level] evidence)

diet



addition of omega-3 fatty acid supplementation to intramuscular progesterone may not further decrease rates of recurrent preterm labor (level 2 [mid-level] evidence) but moderately frequent fish consumption prior to 22 weeks gestation associated with decreased risk (level 2 [mid-level] evidence)  cholesterol-lowering diet may reduce preterm delivery in low-risk pregnancies (level 2 [mid-level] evidence)  bed rest may not be associated with reduced rate of preterm birth (level 2 [mid-level] evidence)  prophylactic beta-mimetic therapy does not appear to reduce rate of preterm delivery in women at increased risk (level 2 [midlevel] evidence)  scaling and root planing for pregnant women with periodontitis may not reduce risk for preterm birth unless periodontal treatment is successful (level 2 [mid-level] evidence)  see Prevention of preterm labor and preterm birth for details Screening American College of Obstetricians and Gynecologists (ACOG) screening recommendations  American College of Obstetricians and Gynecologists (ACOG) screening recommendations  no current data to support the use of fetal fibronectin screening, home uterine activity monitoring, or bacterial vaginosis screening to identify or prevent preterm birth (ACOG Level A)  universal ultrasound screening to determine cervical length may be useful but not mandated in women without a history of preterm birth (ACOG Level B)  practitioners who implement universal cervical length screening should follow 1 of the protocols for transvaginal measurement of cervical length from the available clinical trials (ACOG Level C)  Reference - ACOG Practice Bulletin 130 on prediction and prevention of preterm birth (Obstet Gynecol 2012 Oct;120(4):964), commentary can be found in Obstet Gynecol 2013 Aug;122(2 Pt 1):390 Home uterine activity monitoring (HUAM)  home uterine activity monitoring may decrease risk of preterm birth at < 34 weeks but not perinatal mortality in women at high risk of preterm birth (level 2 [mid-level] evidence)  based on Cochrane review of trials with methodologic limitations  systematic review of 15 randomized or quasi-randomized trials comparing HUAM vs. conventional care or other care packages in 6,008 women at high risk of preterm birth  all trials had ≥ 1 limitation, including  allocation concealment unclear or not stated  lack of or unclear blinding of caregivers  high noncompliance rate  comparing HUAM to standard care  HUAM associated with  nonsignificant decrease in risk of preterm birth at < 37 weeks (risk ratio [RR] 0.85, 95% CI 0.72-1.01) in analysis of 8 trials with 4,834 women, results limited by significant heterogeneity  decreased risk of preterm birth at < 34 weeks in analysis of 3 trials with 1,596 women  RR 0.78 (95% CI 0.62-0.99)  NNT 16-589 with preterm birth at < 34 weeks in 17% of standard care group  increased number of unscheduled antenatal visits (mean difference 0.49 visits, 95% CI 0.36-0.62 visits) in analysis of 2 trials with 3,707 women  increased use of prophylactic tocolysis in analysis of 7 trials with 4,316 women  RR 1.21 (95% CI 1.01-1.45)

 

NNT 12-527 with tocolysis use in 19% of standard care group results not significant in sensitivity analysis restricted to 3 higher-quality trials  no significant differences in  perinatal mortality in analysis of 2 trials with 2,589 infants  preterm birth at < 32 weeks in analysis of 3 trials with 2,550 women  no trials reported maternal anxiety or acceptability  Reference - Cochrane Database Syst Rev 2015 Jan 6;(1):CD006172  increasing frequency of uterine contractions on home uterine activity monitoring may be associated with increased likelihood of preterm delivery < 35 weeks gestation (level 2 [mid-level] evidence)  based on prognostic cohort study without independent validation  306 women with singleton pregnancies 22-24 weeks gestation at increased risk for preterm delivery had 34,908 hours of home monitoring recordings  increasing frequency of uterine contractions associated with increased risk for preterm delivery but no cutoff values had clinically relevant predictive accuracy  for predicting delivery at < 35 weeks  max nighttime contraction frequency ≥ 4/hour at 22-24 weeks gestation had  sensitivity 8.6%  specificity 96.4%  positive predictive value 25%  negative predictive value 88.3%  max daytime contraction frequency ≥ 4/hour at 22-24 weeks gestation had  sensitivity 0%  specificity 98.4%  positive predictive value 0%  negative predictive value 87%  other tests (cervical exam, ultrasound evaluation of cervix, fetal fibronectin assay of cervicovaginal secretions) also had low sensitivity, low positive predictive value, and > 90% negative predictive values  Reference - N Engl J Med 2002 Jan 24;346(4);250 full-text, correction can be found in N Engl J Med 2003 Jul 31;349(5):513, editorial can be found in N Engl J Med 2002 Jan 24;346(4):282, commentary can be found in J Fam Pract 2002 May;51(5):416  review and critique of HUAM for patients in preterm labor can be found in Obstet Gynecol 2008 Aug;112(2 Pt 1):325 Routine vaginal ultrasound in normal pregnancy  insufficient evidence for routine vaginal ultrasound for prevention of preterm birth  insufficient evidence to evaluate routine cervical ultrasound for prevention of preterm birth  based on Cochrane review  systematic review of 5 randomized trials comparing knowledge vs. no knowledge of cervical length via ultrasound in 507 pregnant women at 14-32 weeks gestation  antenatal management evaluated based on knowledge of transvaginal ultrasound cervical length  no trials identified evaluating asymptomatic women with singleton gestations  Reference - Cochrane Database Syst Rev 2013 Jan 31;(1):CD007235  negative screening transvaginal ultrasound at 18-22 weeks gestation may rule out delivery < 35 weeks (level 2 [mid-level] evidence)  based on prognostic cohort study without independent validation  3,694 singleton pregnancies had transabdominal and transvaginal ultrasonography at 18-22 weeks gestation  performance of either cervical length < 29 mm or dilatation of internal cervical os > 5 mm for determining likelihood of delivery < 35 weeks gestation  sensitivity 29%  specificity 96.6%  positive predictive value 6.8%  negative predictive value 99.4%  Reference - Obstet Gynecol 1998 Dec;92(6):902 Vaginal ultrasound to predict recurrence  ultrasound to determine need for cerclage associated with lower cerclage rates and similar pregnancy outcomes compared to history-based cerclage placement in women with history of preterm delivery (level 2 [mid-level] evidence)  based on systematic review limited by clinical heterogeneity  systematic review of 6 studies evaluating pregnancy outcomes following ultrasound-predicated cerclage vs. historypredicated cerclage in 653 women with history of preterm delivery



meta-analysis not performed due to differences in trials including history-based cerclage placed at different gestational ages, varying cervical lengths used as criteria for ultrasound-based cerclage, and differing definitions of cervical insufficiency  no cerclage required in 40%-68% of patients in ultrasound group  5 studies showed no significant difference between groups in preterm delivery rates or pregnancy loss < 24 weeks  1 prospective cohort study found lower rates of preterm delivery < 30 weeks in ultrasound group  Reference - Obstet Gynecol Surv 2008 Dec;63(12):803  cervical ultrasound screening may increase use of cerclage and progesterone supplementation but may not reduce recurrent preterm delivery (level 2 [mid-level] evidence)  based on randomized trial without allocation concealment  247 pregnant women with ≥ 1 previous delivery at 16-34 gestational weeks were randomized to history-indicated cervical cerclage (based on clinician-preference) vs. cerclage for ultrasound finding < 20 mm cervical length and followed through delivery  comparing history-indicated vs. ultrasound-based cerclage  preterm delivery at 24 weeks to 33 weeks plus 6 days occurred in 15% vs. 15% (not significant)  cerclage performed in 19% vs. 32% (p < 0.05)  progesterone supplementation in 25% vs. 32% (p < 0.05)  Reference - CIRCLE trial (Am J Obstet Gynecol 2009 Jun;200(6):623e1 )  vaginal ultrasound with assessment of cervical length may help determine likelihood of recurrence in women with history of preterm delivery  cervical length < 25 mm at 16-19 weeks gestation may be associated with recurrent preterm delivery (level 2 [mid-level] evidence)  based on prognostic cohort study without independent validation  183 women with singleton gestation and history of spontaneous birth before 32 weeks gestation had endovaginal ultrasound every 2 weeks from 16 to 24 weeks gestation  26% had spontaneous preterm birth before 35 weeks gestation  cervical length < 25 mm on initial ultrasound increased risk of preterm birth (relative risk 3.3, 95% CI 2.1-5, 19% sensitivity, 98% specificity, 75% positive predictive value)  dynamic shortening may have added predictive value (p = 0.054)  funneling did not add significant predictive value  serial ultrasound increased sensitivity to 69%, but reduced specificity to 80% (55% positive predictive value)  Reference - JAMA 2001 Sep 19;286(11):1340 full-text  cervical funneling did not independently predict risk of preterm delivery in secondary analysis of this cohort with 183 women and 590 ultrasound scans (Obstet Gynecol 2007 Apr;109(4):863)  second trimester cervical length and funneling may help determine likelihood of preterm delivery, especially in women with history of preterm delivery (level 2 [mid-level] evidence)  based on retrospective prognostic cohort study  2,391 pregnant women at Brazilian teaching hospital had transvaginal ultrasound at 21-24 weeks gestation  analysis included 66 mothers who delivered before 34 weeks (preterm delivery) and 1,892 women who delivered after 34 weeks  mean second trimester cervical length lower in women with preterm delivery (mean 23.8 mm vs. 35.6 mm)  cervical length < 20 mm associated with > 25% risk of preterm delivery  funneling at second trimester ultrasound found in 31 women and associated with earlier delivery (mean 33.5 vs. 38.8 weeks gestation)  significant risk factors for preterm delivery before 34 weeks were  cervical length < 20 mm (odds ratio [OR] 1.12, 95% CI 1.08-1.16)  funneling (OR 6.29, 95% CI 2.52-15.71)  history of preterm delivery (OR 2.71, 95% CI 1.44-5.09)  risk of preterm delivery  7% if second trimester cervical length < 20 mm  34% if second trimester cervical length < 20 mm and funneling  18% if second trimester cervical length < 20 mm and history of preterm delivery  59% if second trimester cervical length < 20 mm, funneling, and history of preterm delivery  Reference - Obstet Gynecol 2005 Mar;105(3):532 Other screening considerations  repeat digital cervical assessment does not detect risk for or reduce rates of preterm birth (level 1 [likely reliable] evidence)  based on Cochrane review











systematic review of randomized trials comparing repeat digital cervical assessment (RDCA) with internal exam or no internal exam (unless clinically indicated)  2 trials with 7,163 women met inclusion criteria  studies with RDCA as a component of multiple tests were not included  no significant difference between groups in preterm birth at < 37 weeks (odds ratio 1.05, 95% CI 0.85-1.31) in analysis of both trials  1 trial with 5,863 women found no significant difference between groups in preterm birth at < 34 weeks, preterm premature rupture of membranes, hospital admission before 37 weeks, cesarean section, use of tocolytic drugs, or neonatal outcomes  insufficient evidence to evaluate adverse effects of digital cervical assessment  Reference - Cochrane Database Syst Rev 2010 Jun 16;(6):CD005940 digital cervical score and Bishop score at 22-29 weeks gestation appear to be relatively poor predictors of spontaneous preterm delivery (level 2 [mid-level] evidence)  based on prognostic cohort study with potential bias in patient population  2,916 women with singleton pregnancy had digital cervical exam at 22-24 weeks gestation and 26-29 weeks gestation to determine Bishop score and digital cervical score  predictive values of Cervical score and Bishop score were estimated based on routinely scheduled exams, and predictive values may be different in high-risk women having exam for nonroutine indications  spontaneous preterm delivery at < 35 weeks occurred in 4.4% of women examined at 22-24 weeks, and in 3.3% of women reexamined at 26-29 weeks  each per-unit increase in digital cervical score associated with decreased risk of preterm birth < 37 weeks (odds ratio 0.62, 95% CI 0.5-0.78)  each per-unit increase in Bishop score associated with increased risk of spontaneous preterm delivery < 37 weeks (odds ratio 1.46, 95% CI 1.29-1.65)  authors conclude low event prevalence limits predictive capability of these tests  Reference - Obstet Gynecol 2008 Sep;112(3):508 full-text fetal fibronectin plus short cervical length may aid in prediction of preterm birth in women with symptoms of preterm labor (level 2 [mid-level] evidence)  based on systematic review limited by clinical heterogeneity  systematic review of 8 cohort studies and 1 randomized trial evaluating fetal fibronectin plus short cervical length for prediction of preterm birth in women with symptoms of preterm labor  all studies used fetal fibronectin cutoff ≥ 50 ng/mL  definition of short cervix not uniform across studies  most studies enrolled women with symptoms of preterm labor at 22-35 weeks gestation  preterm birth in  3.6% (95% CI 1.6%-7.1%) at < 7 days of testing in analysis of 2 studies with 192 women  4.4% (95% CI 2.2%-8%) at ≤ 14 days of testing in analysis of 2 studies with 203 women  performance of fetal fibronectin plus short cervical length for prediction of preterm birth  at < 7 days (2 studies)  sensitivity 71.4% (95% CI 35.9%-91.8%)  specificity 96.8% (95% CI 93.1%-98.5%)  at ≤ 14 days (2 studies)  sensitivity 33.3% (95% CI 12.1%-64.6%)  specificity 86% (95% CI 80.3%-90.3%)  Reference - Am J Obstet Gynecol 2013 Mar;208(3):233.e1 in women with short cervix, positive fetal fibronectin test results may not identify women for whom cerclage would prevent spontaneous preterm birth (level 2 [mid-level] evidence)  based on secondary analysis of randomized trial  217 pregnant women (mean age 27-29 years) with short, funneled cervix at 18-24 weeks gestation had fetal fibronectin (fFN) test  29% were fFN-positive and 71% were fFN-negative  57% had cerclage and 43% had no cerclage  no significant differences in preterm birth rates comparing  fFN-positive women with vs. without cerclage (44.1% vs. 55.2%)  fFN-negative women with vs. without cerclage (17.8% vs. 17.2%)  Reference - Am J Obstet Gynecol 2009 Feb;200(2):158e1 positive fFN test and/or cervical length < 20 mm associated with increased risk of preterm delivery in asymptomatic twin pregnancies (level 2 [mid-level] evidence)

  



based on retrospective prognostic cohort study 155 asymptomatic twin pregnancies with fFN and cervical length tested at 22-32 weeks gestation evaluated positive fFN test or cervical length < 20 mm each independently associated with significantly increased risk of preterm delivery at < 28, < 30, < 32, < 34, and < 37 weeks gestation  combination of positive fFN and cervical length < 20 mm had higher positive predictive value for preterm delivery at all time points compared with either positive test alone  Reference - Am J Obstet Gynecol 2009 Sep;201(3):313e1 low salivary progesterone (< 2,575 pg/mL) might be associated with early preterm birth in asymptomatic pregnant women at high risk for preterm birth (level 2 [mid-level] evidence)  based on prognostic cohort study without independent validation  90 asymptomatic pregnant women at high risk for preterm delivery were evaluated for salivary progesterone concentration at 24-28 weeks gestation and 3-4 weeks later  salivary progesterone level < 2,575 pg/mL predicted delivery at < 34 weeks gestation with  sensitivity 83%  specificity 86%  positive predictive value 60%  negative predictive value 95%  Reference - BJOG 2013 Jul;120(8):1003