Cahill, Alison G. MD, MSCI; Caughey, Aaron B. MD, PhD; Roehl, Kimberly A. MPH; Odibo, Anthony O. MD, MSCE; Macones, George A. MD, MSCE
Fetal heart rate decelerations during the second stage of labor at term are frequent.1 However, the incidence of terminal decelerations, prolonged fetal heart rate decelerations that precede and are unresolved before delivery, is unknown. Additionally, the association between length of terminal deceleration and acidemia during typical labor is also unknown. This leaves clinicians without specific evidence to guide their clinical decision-making regarding the allowable length of terminal deceleration before needing to intervene with alterations in second-stage management or operative delivery. Currently, outcomes related to terminal decelerations must be extrapolated from published evidence in two specific patient populations, laboring women undergoing cesarean delivery and women attempting vaginal birth after cesarean delivery. Among women who undergo cesarean delivery, there are data describing the association between decision-to-incision time for operative intervention and outcomes.2–4 Published data among patients attempting vaginal birth after cesarean delivery describing the relationship between duration from uterine rupture to delivery and risk of acidemia and adverse outcomes have described duration of 18 minutes or more to be associated with increased risk of adverse outcome.5,6 However, rare and acute clinical events such as uterine rupture, which are known to be independently associated with an a priori increase in risk of acidemia, may not be generalizable to the terminal decelerations seen in most patients.
Specific knowledge of the relationship between terminal deceleration duration and risk of acidemia could inform the decision to perform a cesarean delivery or operative vaginal delivery, as well as the urgency with which to proceed. The goal of this study was to estimate the incidence of terminal decelerations during the second stage of term labor and to estimate the association between terminal decelerations and their characteristics and acidemia and markers of neonatal depression.
MATERIALS AND METHODS
This was a retrospective cohort study of all consecutive births between 2004 and 2008 at Washington University in St. Louis Medical Center. Women included in the study carried a singleton, vertex gestation at term (at or after 37 0/7 weeks), labored, and reached complete dilation. Women were excluded if they carried a multifetal gestation, if had a fetus with a known congenital anomaly, or if they did not have sufficient electronic fetal monitoring (EFM) recording during the 30 minutes before delivery (defined as less than 10 minutes of EFM during the 30 minutes before delivery). The institutional policy is one of universal EFM during labor and arterial umbilical cord gas pH level at delivery. The study was conducted after approval from the Washington University School of Medicine Human Research Protection Office.
Two formally trained obstetric research nurses certified in EFM interpretation and blinded to clinical data and outcomes extracted the 30 minutes of EFM before delivery. Electronic fetal monitoring was extracted using the Eunice Kennedy Shriver National Institute of Child Health and Human Development and the American College of Obstetricians and Gynecologists three-tiered category system and definitions for decelerations to extract the EFM.7 Specific to this study, the presence of a terminal deceleration was primarily extracted, as were features of the deceleration, including duration, nadir, and variability within the nadir. The primary comparison was between women with a terminal deceleration, defined as a prolonged deceleration (15 bpm or more below baseline for 120 seconds or more and fewer than 10 minutes) or bradycardia (less than 110 bpm for 10 minutes or more) immediately preceding delivery, and those without a terminal deceleration.8 If tracing was lost or discontinuous after the initiation of the terminal deceleration, it was assumed that the duration was until delivery to bias any potential effect on the results toward the null hypothesis. Interval interobserver reliability was performed, which has been previously described as high.1 For presence of terminal decelerations, kappa coefficient was consistently more than 0.9. Detailed maternal and pregnancy data were also extracted. Data included medical and obstetric history, pregnancy course and complications, medication exposure, labor course and acute events (including placental abruption, umbilical cord prolapse, and uterine rupture), physical examination, anesthesia type, delivery, and neonatal outcomes. Diagnoses of placental abruption, uterine rupture, and cord prolapse were made clinically, and a composite of these three outcomes was made to represent a subgroup (abrupt composite) with a priori risk for adverse outcomes. Use of internal monitors for fetal heart rate monitoring and contractions and umbilical cord gas arterial pH level, as well as CO2 and base excess, also were recorded. The primary outcome was acidemia, defined as arterial umbilical cord gas pH level of 7.10 or less. Secondary outcomes included arterial umbilical cord gas pH level 7.05 or less, base excess more than −8, metabolic acidemia (pH level 7.10 or less and base excess more than −8), admission to the neonatal intensive care unit (level IV) or admission to the special care unit (level II), and Apgar score less than 7 at 5 minutes.
The cohort was described and baseline characteristics were compared between women who had a terminal deceleration and those who did not. Independent Student t tests and Mann-Whitney U tests were used for continuous variables and χ2 and Fisher exact tests were used for dichotomous variables as appropriate. Continuous variables were tested for normality with the Shapiro-Francia test.9 Relative risk of acidemia and 95% confidence intervals (CIs) were calculated for each of the outcomes of interest. 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 terminal decelerations and acidemia by eliminating nonsignificant factors. The final model, adjusting only for nulliparity, was tested with the Hosmer-Lemeshow goodness-of-fit test. Among women with a terminal deceleration, characteristics of the terminal deceleration were described by acidemia as well as by mode of delivery. A secondary analysis was performed to explore the risk of acidemia and other adverse outcomes among women with terminal bradycardia. Linear regression was then used to estimate the incremental association between increasing terminal deceleration duration beyond 2 minutes and decreasing arterial umbilical cord pH level. Receiver-operator characteristic curve analysis was used to estimate the predictive ability of terminal deceleration duration and risk of acidemia, and test characteristics of bradycardia for acidemia were estimated. All analyses were performed using STATA 10 special edition.
Of 5,388 women meeting inclusion criteria (Fig. 1), 951 (17.7%) experienced a terminal deceleration and 4,437 (82.3%) did not. The groups were similar with respect to maternal age and race, body mass index, gestational age at delivery, use of regional anesthesia, and labor type (Table 1). Women experiencing a terminal deceleration were more likely to be nulliparous and, as expected, they were less likely to have a spontaneous vaginal delivery. Finally, the cohort overall tended to be obese.
Overall, presence of a terminal deceleration was not significantly associated with an arterial umbilical cord pH level of 7.10 or less (adjusted odds ratio [OR] 1.2; 95% CI 0.6–2.3) or 7.05 or less (adjusted OR 1.4; 95% CI 0.5–4.4) (Table 2). Terminal decelerations similarly were not associated with an increased risk of Apgar score less than 7 at 5 minutes (adjusted OR 0.4; 95% CI 0.1–1.1) or admission to a level II or level IV nursery (adjusted OR 0.8; 95% CI 0.6–1.2). However, terminal decelerations were more common in the setting of the abruption composite (1.1% compared with 0.4%; adjusted OR 3.0; 95% CI 1.4–6.7).
Among the 951 women with a terminal deceleration, only 12 (1.3%) delivered a newborn with acidemia. The median duration of the terminal deceleration among the entire cohort was 3.3 minutes, and this differed by acidemia (Table 3). Terminal decelerations were longer on average among newborns with acidemia (6.7 minutes compared with 3.2 minutes; P<.01). For every additional 120 seconds of duration of the terminal deceleration beyond the first 120 seconds, there was a corresponding decrease in arterial umbilical cord pH level by 0.042 (95% CI 0.040–0.048; P<.01). However, terminal deceleration characteristics, such as median or greatest depth and variability within the nadir, were not associated with risk of acidemia (Table 3). Duration of terminal deceleration beyond 2 minutes had a modest predictive ability for acidemia (pH level of 7.10 or less; area under the curve, 0.78; 95% CI 0.60–0.94) (Fig. 2). Although there was no clear inflection point, the most favorable test characteristics came from a cut-off of 4 minutes or more, which had a sensitivity of 75.0% (95% CI 74.2–76.3%) and a specificity of 64.0% (95% CI 62.8–65.1%) for acidemia.
Among those with a terminal deceleration, 31 women had a bradycardia (duration of 10 minutes or more) as compared with 930 with a terminal deceleration less than 10 min. Bradycardia was associated with an increased risk of acidemia (pH level of 7.05 or less), increased base excess, and metabolic acidemia (pH level of 7.10 or less and base excess more than −8). Bradycardia also increased the risk for level II or level IV nursery admission (adjusted OR 3.8; 95% CI 1.4–10.7) (Table 4), which remained even when women experiencing the abruption composite were excluded (adjusted OR 3.7; 95% CI 1.3–10.4). Despite its association, presence of bradycardia (10 minutes or more) was poorly predictive of acidemia, with a sensitivity of 33.3%, a specificity of 97.0%, and a positive predictive value of only 12.9%.
We found terminal decelerations, defined as a prolonged deceleration of 2 minutes or more before delivery, to be common in term deliveries from the second stage. We found no association between overall presence of a terminal deceleration and risk of arterial umbilical cord pH level of 7.10 or less or 7.05 or less, base excess more than −8 or more than −12, Apgar score less than 7 at 5 minutes, or admission to a level II or level IV nursery. Among women with a terminal deceleration, we found that more than 98% experience delivery of a newborn with a pH level more than 7.10, and features of the deceleration such as depth and variability were not associated with acidemia.
However, we found that neonates born with acidemia were more likely to have longer median terminal decelerations, and that there was a significant association between increasing terminal deceleration duration and decreasing arterial umbilical blood gas pH level. Length of terminal deceleration beyond 2 minutes demonstrated some predictive ability for pH level of 7.10 or less, but it was not sufficiently robust as a clinical test to discriminate between newborns with and without acidemia. Finally, among women with a terminal deceleration, bradycardia was associated with increased risks of acidemia, elevated base excess, and level II or level IV nursery admission, even if those experiencing an acute clinical event such as placental abruption, cord prolapse, or uterine rupture were excluded.
Fetal heart rate pattern characteristics are often used clinically for management decisions during the second stage of labor, such as changes in instructions for maternal pushing or decisions for operative delivery. Several authors have published data regarding the duration between decision for operative delivery and outcomes. Bloom et al2 published a prospective cohort study of more than 11,000 women undergoing primary cesarean delivery, specifically examining the 2,808 undergoing emergency cesarean delivery. Comparing those who underwent emergency cesarean delivery within 30 minutes of the decision with those who delivered beyond 30 minutes, the most common outcome among the entire cohort was a newborn without evidence of depression (95%); there was no increase in neonatal depression, acidemia, or neonatal depression in those delivered beyond 30 minutes, similar to the findings previously reported. Unfortunately, although the majority of cesarean deliveries were performed because of nonreassuring fetal heart rate patterns, data were not provided regarding specific characteristics of the patterns. Specifically, no data regarding prolonged terminal decelerations or bradycardia were provided, making it difficult to extrapolate these findings for clinical decision-making in the setting of these second-stage patterns.
We previously have described overall EFM patterns during the last 30 minutes before delivery during the second stage as universally overall category II.1 Thus, specific periodic changes within category II patterns, such as terminal decelerations, likely have the greatest effect on bedside clinical decision-making. Kamoshita et al10 performed a retrospective cohort study of emergency cesarean deliveries for bradycardia, which they defined as less than 100 bpm lasting for more than 5 minutes. Among the 19 women studied, they found that the onset of bradycardia to delivery interval was negatively correlated with umbilical artery pH level, but that an interval of 25 minutes or less resulted in normal neonatal outcomes, leading the authors to conclude that delivery within 25 minutes of the onset of bradycardia improves outcomes. However, the 19 patients who met inclusion criteria for bradycardia limited the analytic approach and generalizability of the findings. Sheiner et al11 described the clinical significance of abnormal second-stage fetal heart rate tracings in 420 patients, reporting an association between fetal bradycardia less than 70 bpm before delivery and an increased risk of pH level less than 7.20 and base excess more than 12 mmol/L. The authors concluded that these patterns might jeopardize fetal well-being, and thus expedited delivery should be considered. However, the small sample size with bradycardia limited the precision of their estimates and their ability to interrogate characteristics of the decelerations.
In the specific setting of uterine rupture, Holmgren et al5 found that intervals from rupture to delivery more than 18 minutes were associated with increased risk of acidemia and neurologic injury, consistent with findings reported previously.6 However, although variable decelerations and bradycardia were associated with uterine rupture, they were not seen in all cases. Further, given that uterine rupture itself is associated with adverse neonatal outcomes such as acidemia, the generalizability of these findings to women with terminal decelerations in the absence of such an event is limited. The present study offers evidence to fill this knowledge gap.
A unique strength of this cohort study was the interpretation of the fetal heart rate tracing deceleration characteristics by two dedicated obstetric research nurses formally trained in EFM interpretation who were blind to all clinical factors and outcome data, including arterial umbilical cord gas pH level. Although some may argue that human interpretation can be unreliable, these nurses underwent formal interobserver and intraobserver reliability testing and ongoing retraining, and they demonstrated high reproducibility. Further, human interpretation is how EFM is used in everyday practice, and thus offers generalizability. We were able to adjust for potentially confounding factors given our large sample size, and we were able to investigate characteristics of the terminal deceleration with acidemia and outcome, including duration and depth.
In contrast, there are some potential limitations that are important to consider with respect to our study. Choosing the 30 minutes before delivery may limit generalizability of our findings, specifically in the interpretation of atypical deceleration characteristics at other times during labor. However, we chose this time period because it is most proximal to the measurement of acidemia in modern practice (at birth) and few terminal decelerations are allowed the natural history of more than 30 minutes duration. Although we chose to define our terminal decelerations as prolonged or bradycardia before delivery, and although we chose to explore duration beyond 10 minutes and risk of acidemia, our study certainly was not exhaustive and there are other possible features of terminal decelerations that might be associated with acidemia but have yet to be tested. Finally, despite our large sample size, the rarity of acidemia in nonanomalous term newborns limited our power to detect between-group differences, particularly among those with terminal decelerations.
Despite these potential limitations, we believe that this study adds to the literature. We found that terminal decelerations are common in term newborns delivered from the second stage, and that more than 98% of them will be delivered with a normal umbilical cord arterial pH level. However, we also found an association between increasing duration of terminal deceleration beyond the initial 120 seconds and decreasing pH level. Although the duration of a terminal deceleration alone cannot efficiently predict acidemia, there is an association between increasing terminal deceleration duration and risk of decreasing pH level, as well as higher-level nursery admission and acidemia in those with bradycardia (duration 10 minutes or more). This information can be incorporated into clinical decision-making regarding urgency of delivery.
1. Cahill AG, Roehl KA, Odibo AO, Macones GA. Association and prediction of neonatal acidemia. Am J Obstet Gynecol 2012;207:206.e1–8.
2. Bloom SL, Leveno KJ, Spong CY, Gilbert S, Hauth JC, Landon MB, et al.. Decision-to-incision times and maternal and infant outcomes. Obstet Gynecol 2006;108:6–11.
3. Chauhan SP, Roach H, Naef RW II, Magann EF, Morrison JC, Martin JN Jr. Cesarean section for suspected fetal distress. Does the decision-incision time make a difference? J Reprod Med 1997;42:347–52.
4. Nasrallah FK, Harirah HM, Vadhera R, Jain V, Franklin LT, Hankins GD. The 30-minute decision-to-incision interval for emergency cesarean delivery: fact or fiction? Am J Perinatol 2004;21:63–8.
5. Holmgren C, Scott JR, Porter TF, Esplin MS, Bardsley T. Uterine rupture with attempted vaginal birth after cesarean delivery: decision-to-delivery time and neonatal outcome. Obstet Gynecol 2012;119:725–31.
6. Leung AS, Leung EK, Paul RH. Uterine rupture after previous cesarean delivery: maternal and fetal consequences. Am J Obstet Gynecol 1993;169:945–50.
7. Intrapartum fetal heart rate monitoring: nomenclature, interpretation, and general management principles. ACOG Practice Bulletin No. 106. American College of Obstetricians and Gynecologists. Obstet Gynecol 2009;114:192–202.
8. Macones GA, Hankins GD, Spong CY, Hauth J, Moore T. The 2008 National Institute of Child Health and human development workshop report on electronic fetal monitoring: update on definitions, interpretation, and research guidelines. Obstet Gynecol 2008;112:661–6.
9. Royston P. A pocket-calculator algorithm for the Shapiro-Francia test for non-normality: an application to medicine. Stat Med 1993;12:181–4.
10. Kamoshita E, Amano K, Kanai Y, Mochizuki J, Ikeda Y, Kikuchi S, et al.. Effect of the interval between onset of sustained fetal bradycardia and cesarean delivery on long-term neonatal neurologic prognosis. Int J Gynaecol Obstet 2010;111:23–7.
11. Sheiner E, Hadar A, Hallak M, Katz M, Mazor M, Shoham-Vardi I. Clinical significance of fetal heart rate tracings during the second stage of labor. Obstet Gynecol 2001;97(5 Pt 1):747–52.
Figure. No available...Image Tools
© 2013 by The American College of Obstetricians and Gynecologists.