Objective
The purpose of this study was to identify electronic fetal monitoring patterns that are associated with neonatal respiratory morbidity.
Study Design
In an on-going prospective cohort study of >8000 consecutive term, vertex, nonanomalous singleton pregnancies during labor, we performed this analysis within the first 5000 women as a representative sample. Electronic fetal monitoring patterns in the 30 minutes preceding delivery were extracted by trained obstetrics research nurses, who were blinded to clinical data, using the National Institute of Child Health and Human Development system; the data were compared between those with respiratory morbidity and healthy infants (no morbidities). The primary outcome was neonatal respiratory morbidity, which was defined as either oxygen requirement at ≥6 hours of life or any mechanical ventilation in the first 24 hours. Multivariable logistic regression was used to adjust for confounders.
Results
Of 4736 neonates, 175 (3.4%) experienced respiratory morbidity. Most electronic fetal monitoring patterns were category II (96.6%; n = 4575). Baseline tachycardia (adjusted odds ratio [aOR], 2.9; 95% confidence interval [CI], 1.9–4.4), marked variability (aOR, 2.7; 95% CI, 1.5–5.0), and prolonged decelerations (aOR,2.7; 95% CI, 1.5–5.0) were significantly associated with an increased likelihood of term neonatal respiratory morbidity. Accelerations and persistent moderate variability were both significantly associated with a decreased likelihood of respiratory morbidity.
Conclusion
Specific features of category II electronic fetal monitoring patterns make respiratory morbidity more likely in nonanomalous term infants. Tachycardia, marked variability, or prolonged decelerations before delivery can assist providers in anticipating the potential need for neonatal respiratory support.
Respiratory morbidity is estimated to occur in 6.0-8.8% of early term (37-38 weeks of gestation) and 2.1-4.7% of term (≥39 weeks of gestation) infants. However, despite being relatively uncommon, it remains the most common morbidity among these infants and has the potential to cause significant morbidity and even death. Secondary complications may include persistent pulmonary hypertension of the neonate, severe hypoxic respiratory failure, bronchopulmonary dysplasia, pneumothorax, and neurologic complications. Infants may require treatment with nitric oxide, mechanical ventilation, or extracorporeal membrane oxygenation and subsequently may experience adverse outcomes because of invasive procedures. Despite its significance, we lack the tools to identify which infants born after 37 weeks of gestation are at risk for respiratory morbidity and may benefit from additional neonatal support at delivery.
Currently, there are few intrapartum risk factors for neonatal respiratory morbidity in nonanomalous term infants. Given the ubiquitous use of intrapartum electronic fetal monitoring (EFM) and that the most common significant morbidity in these term infants is respiratory, we sought to determine which characteristics of EFM were associated with an increase in neonatal respiratory morbidity. Specifically, we assessed the associations between fetal bradycardia or tachycardia, absent, minimal, moderate, or marked variability, presence of accelerations, and presence of early, variable, late, or prolonged decelerations, and neonatal respiratory morbidity.
Materials and Methods
We conducted this study within an on-going prospective cohort study of >8000 consecutive term, vertex, nonanomalous singleton pregnancies during labor at Washington University in St. Louis, MO; this study included the first 5000 subjects as a representative sample. Neonates with <10 minutes of EFM in the 30 minutes before delivery, who were <37 weeks’ gestational age, and who had a postnatal anomaly diagnosis were excluded. The institutional policy is one of universal EFM during labor. The study was conducted after approval from the Washington University School of Medicine Human Research Protection Office (IRB #201107131, approved 12/11/2014).
Obstetric research nurses who had been formally trained in EFM pattern recognition were blinded to the antepartum and neonatal clinical data and extracted EFM patterns in the 2 hours before delivery. These patterns were categorized according to fetal heart rate baseline; variability, presence, and number of accelerations; and presence, number, and type of decelerations. We compared characteristics of EFM patterns using the Eunice Kennedy Shriver National Institute of Child Health and Human Development 3-tiered category system between infants with respiratory morbidity and healthy infants without morbidities. Maternal and neonatal demographic data that included obstetric and gynecologic history, type of labor, types of augmentation used, mode of delivery, maternal complications, use of regional anesthesia, and neonatal birthweight were also obtained.
In this study, we examined the EFM patterns in the 30 minutes before delivery. EFM was performed with the use of internal or external monitoring as clinically indicated. The primary outcome was neonatal respiratory morbidity, which was defined as either any oxygen requirement at or after 6 hours of life or any mechanical ventilation in the first 24 hours. Because cesarean delivery and fever are both risk factors for increased neonatal respiratory morbidity, 2 secondary analyses were performed that excluded those patients who underwent cesarean delivery and those with fever. Because mechanical ventilation is the most severe acute respiratory morbidity for a term infant, analyses were repeated to estimate which EFM patterns were associated with mechanical ventilation compared with those without morbidity.
Baseline characteristics of women who delivered infants with and without respiratory morbidity were compared. The Student t test and Mann-Whitney U test were used for continuous variables; χ 2 and Fisher exact tests were used for dichotomous variables as appropriate. Continuous variables were tested for normality with the Shapiro-Francia test. Relative risks of severe respiratory morbidity and 95% confidence intervals were calculated for each of the EFM characteristics. Stratified analyses were performed to identify potentially confounding factors, which were considered in multivariable analyses. Multivariable logistic regression was performed in a backward step-wise fashion to refine estimates of association between EFM characteristics and neonatal respiratory morbidity by controlling for confounding factors. Model fit of the final model (adjustment for maternal fever, parity, pregestational diabetes mellitus, previous cesarean delivery, and preeclampsia) was tested with the Hosmer-Lemeshow goodness-of-fit test. We included all subjects who met inclusion criteria during the study period; no a priori sample size estimation was performed. All analyses were performed using STATA 10 special edition (StataCorp, College Station, TX) and SAS software (version 9.2; SAS Institute Inc, Cary, NC).
Results
Of 4736 neonates who met the inclusion criteria, 175 (3.7%) experienced respiratory morbidity, and 4561 (96.3%) did not ( Table 1 ). Most EFM patterns were category II (96.6%; n = 4575) during the 30 minutes before delivery. Maternal age, gestational age at delivery, labor type, birthweight, percentage of maternal black race, percentage of gestational diabetes mellitus, and use of regional anesthesia, Foley bulb, and oxytocin were not significantly different between the groups. As expected, neonates with respiratory morbidity were less likely to deliver vaginally and more likely to deliver by cesarean. Preeclampsia, pregestational diabetes mellitus, nulliparity, previous cesarean delivery, higher body mass index, prostaglandin use, and maternal fever were significantly more likely among women who delivered an infant with respiratory morbidity.
| Characteristic | Respiratory morbidity (n = 175) | No respiratory morbidity (n = 4561) | P value |
|---|---|---|---|
| Maternal age, y a | 24.9 ± 6.4 | 25.6 ± 5.9 | .13 |
| Advanced maternal age, % | 5.7 | 8.5 | .19 |
| Gestational age at delivery, wk a | 38.9 ± 1.3 | 38.9 ± 1.2 | .55 |
| Maternal black race, % | 66.3 | 65.0 | .72 |
| Body mass index, kg/m 2 a | 33.0 ± 7.1 | 31.9 ± 7.3 | .05 |
| Preeclampsia, % | 16.0 | 9.4 | < .01 |
| Gestational diabetes mellitus, % | 3.4 | 2.7 | .48 |
| Pregestational diabetes mellitus, % | 4.0 | 0.9 | < .01 |
| Nulliparous, % | 57.4 | 40.2 | < .01 |
| Previous cesarean delivery, % | 16.0 | 8.4 | < .01 |
| Labor type, % | .16 | ||
| Spontaneous | 25.1 | 30.0 | |
| Augmented | 24.6 | 26.8 | |
| Induction | 50.3 | 43.2 | |
| Regional anesthesia, % | 89.1 | 89.5 | .90 |
| Prostaglandin, % | 24.0 | 17.2 | .02 |
| Foley bulb, % | 12.0 | 9.1 | .18 |
| Oxytocin, % | 66.9 | 65.9 | .79 |
| Birthweight, g a | 3260 ± 548 | 3238 ± 453 | .53 |
| Birthweight >4000 g, % | 6.3 | 5.0 | .48 |
| Vaginal delivery, % | 50.3 | 80.1 | < .01 |
| Operative vaginal delivery, % | 6.9 | 5.3 | .39 |
| Cesarean delivery, % | 42.9 | 14.6 | < .01 |
| Maternal fever, % | 16.6 | 1.3 | < .01 |
Neonates who experienced respiratory morbidity were significantly more likely to ever have had baseline tachycardia (adjusted odds ratio [aOR], 2.9; 95% confidence interval [CI], 1.9–4.4) or marked variability (aOR, 2.7; 95% CI, 1.5–5.0) on EFM tracings in the 30 minutes before delivery compared with healthy neonates ( Table 2 ). Prolonged decelerations (aOR, 1.7; 95% CI, 1.3–2.4) were also significantly more likely in neonates with respiratory morbidity. The associations remained significant after adjustment for fever, parity, pregestational diabetes mellitus, previous cesarean delivery, preeclampsia, and labor type. However, the presence of early (aOR, 0.4; 95% CI, 0.1–1.1), variable (aOR, 0.8; 95% CI, 0.5–1.1), or late (aOR, 0.8; 95% CI, 0.6–1.1) decelerations were not associated independently with term neonatal respiratory morbidity. Similarly, baseline bradycardia (aOR, 0.5; 95% CI, 0.1–3.4) and ever (aOR, 1.3; 95% CI, 0.9–1.8), mostly (aOR, 1.1; 95% CI, 0.8–1.6), and always (aOR, 1.2; 95% CI, 0.8–1.7) absent, or minimal variability were not associated independently with respiratory morbidity in the term neonate. Moderate variability throughout the final 30 minutes before delivery and the presence of accelerations were both associated with a decreased likelihood of respiratory morbidity in the term neonate (aOR, 0.7; 95% CI, 0.5–0.9 and aOR, 0.6; 95% CI, 0.4–0.9, respectively).
| Characteristic a | Respiratory morbidity | No respiratory morbidity | Relative risk (95% confidence interval) | Adjusted odds ratio (95% confidence interval) |
|---|---|---|---|---|
| Baseline, % | ||||
| Ever bradycardia <110 beats/min | 0.7 | 1.9 | 0.37 (0.05–2.63) | 0.5 (0.1–3.4) b |
| Ever baseline <120 beats/min | 8.2 | 13.4 | 0.59 (0.33–1.06) | 0.7 (0.4–1.3) b |
| Ever tachycardia >160 beats/min | 29.0 | 7.6 | 4.46 (3.20–6.23) | 2.9 (1.9–4.4) c |
| Variability, % | ||||
| Ever absent or minimal | 62.0 | 53.9 | 1.38 (1.01–1.89) | 1.3 (0.9–1.8) c |
| Mostly absent or minimal | 43.2 | 38.9 | 1.19 (0.87–1.63) | 1.1 (0.8–1.6) c |
| Always absent or minimal | 34.8 | 28.9 | 1.30 (0.94–1.80) | 1.2 (0.8–1.7) c |
| Mostly moderate | 47.1 | 56.3 | 0.70 (0.51–0.95) | 0.7 (0.5–1.0) c |
| Always moderate | 33.6 | 43.7 | 0.66 (0.48–0.92) | 0.7 (0.5–0.9) c |
| Ever marked | 8.7 | 3.2 | 2.69 (1.56–4.64) | 2.7 (1.5–5.0) c |
| Accelerations | ||||
| Present, % | 22.2 | 33.6 | 0.57 (0.39–0.84) | 0.6 (0.4–0.9) c |
| n, median (interquartile range) | 2 (1–3) | 2 (1–3) | — | — |
| Decelerations | ||||
| Present, % | 78.2 | 86.1 | 0.60 (0.42–0.85) | 0.8 (0.5–1.2) c |
| n, median (interquartile range) | 8 (3–10) | 6 (3–9) | — | — |
| Early, % | 2.4 | 6.7 | 0.34 (0.13–0.92) | 0.4 (0.1–1.1) c |
| Variable, % | 71.2 | 80.7 | 0.61 (0.44–0.84) | 0.8 (0.5–1.1) c |
| Late, % | 41.8 | 50.1 | 0.72 (0.54–0.98) | 0.8 (0.6–1.1) c |
| Prolonged, % | 45.9 | 33.7 | 1.64 (1.22–2.20) | 1.7 (1.3–2.4) c |
a Always was defined as the presence of the characteristic during the entire 30-minute period before delivery; Ever and Mostly refer to the presence of the electronic fetal monitoring characteristic during any 10-minute segment and the presence of the electronic fetal monitoring characteristic for any ≥15-minute segment, respectively, in the 30-minute period before delivery
b Adjusted for fever, parity, pregestational diabetes mellitus, previous cesarean delivery, preeclampsia, labor type
c Adjusted for fever, parity, pregestational diabetes mellitus, previous cesarean delivery, preeclampsia.
The secondary analyses that excluded women who underwent cesarean delivery ( Table 3 ) and women who experienced maternal fever ( Table 4 ) revealed consistent findings with baseline tachycardia, marked variability, and prolonged decelerations significantly more likely to precede the delivery of a neonate that experienced respiratory morbidity. Analysis that included only neonates who were ventilated mechanically showed an independent association of baseline tachycardia and prolonged decelerations with term neonatal respiratory morbidity, but not for marked variability ( Table 5 ). In all 3 secondary analyses, baseline bradycardia, absent or minimal variability, and early, variable, and late decelerations were not associated with neonatal respiratory morbidity, with the exception of variable decelerations when women who underwent cesarean delivery were excluded.
| Characteristic a | Respiratory morbidity (n = 100), % | No respiratory morbidity (n = 3894), % | Relative risk (95% confidence interval) | Adjusted odds ratio b (95% confidence interval) |
|---|---|---|---|---|
| Ever baseline tachycardia >160 beats/min | 27.1 | 6.8 | 4.7 (3.1–7.3) | 3.0 (1.8–5.1) |
| Always moderate variability | 35.0 | 43.7 | 0.7 (0.5–1.0) | 0.7 (0.5–1.1) |
| Ever marked variability | 9.2 | 3.2 | 2.9 (1.5–5.6) | 2.7 (1.3–5.7) |
| Accelerations present | 27.6 | 34.5 | 0.7 (0.5–1.1) | 0.8 (0.5–1.2) |
| Variable decelerations | 95.9 | 87.5 | 3.3 (1.2–8.9) | 3.4 (1.2–9.5) |
| Prolonged decelerations | 52.0 | 35.7 | 1.9 (1.3–2.8) | 1.8 (1.2–2.8) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
