Induction of labor is performed for 24.5% of all births in the United States.1 Indications for induction include maternal comorbidities (eg, diabetes mellitus, hypertensive diseases) and conditions that affect fetal wellbeing (eg, prelabor rupture of membranes, fetal growth restriction).2 Typically, an induction of labor starts with cervical ripening followed by continuous infusion of synthetic oxytocin to promote regular, repetitive contractions. Occasionally, despite prolonged exposure to oxytocin, a regular contraction pattern is not attained, or cervical change does not occur. In these circumstances, some clinicians temporarily discontinue the oxytocin infusion (“oxytocin rest”) with the theory that it will re-sensitize the myocyte's response to oxytocin.
The practice of oxytocin rest has biological plausibility. As with other G protein-coupled receptors, oxytocin receptors on myocytes become saturated after prolonged oxytocin exposure and are internalized and degraded.3,4 Oxytocin receptor mRNA becomes down-regulated.5,6 Consequently, the uterine response to additional oxytocin treatment diminishes.4,7–10
Despite this evidence, the clinical efficacy of oxytocin rest during induction of labor remains unclear. Evidence examining its use in latent labor is limited. Complete cessation of oxytocin during the active phase of labor has no effect on mode of delivery.11 Our objective was to assess whether oxytocin rest in a protracted latent phase during induction of labor, here defined as 8 hours of oxytocin without entering active phase, is associated with the mode of delivery among parturients and, if so, to identify the duration of oxytocin rest associated with the lowest risk of cesarean delivery.
This is a retrospective cohort study of patients who underwent induction of labor with intravenous oxytocin infusion at one large tertiary medical center (Yale New Haven Hospital) with two campuses from January 1, 2012, through November 30, 2016. This study was approved by the Yale University Human Investigation Committee.
Nulliparous women with singleton, vertex pregnancies who underwent induction of labor at term were identified from the electronic medical record (EMR). Patient characteristics including age, race or ethnicity, body mass index (BMI, calculated as weight in kilograms divided by height in meters squared) at time of admission, gestational age at the time of delivery, and neonatal birth weight were extracted from the EMR. Maternal comorbidities and obstetric risk factors were identified using International Classification of Diseases (ICD) codes (Appendix 1, available online at http://links.lww.com/AOG/B730). The labor flowsheet that is maintained by the patient's nurse during the course of her induction was used to assess the timing of the patient's oxytocin initiation and dosing, cervical dilation throughout her labor course, and the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) fetal heart rate category at the time of oxytocin rest. Indications for induction of labor were incompletely coded in the EMR, so for this analysis, medical or obstetric complications associated with the admission were used as the indication. For patients who did not have a medical or obstetric complication but who were at 41 weeks of gestation or greater, the indication for induction was listed as “late term or postterm.” Remaining patients were listed as being induced for “other or elective.”
At our institution, patients who present with an unripe cervix first undergo cervical ripening before initiation of oxytocin. Intrauterine balloons are most frequently used; pharmacologic agents (vaginal misoprostol) are also used when an intrauterine balloon cannot be placed. During the years comprising this analysis, nulliparous patients rarely underwent concurrent mechanical ripening and oxytocin infusion.
At our institution, the administration of oxytocin for induction of labor is protocol driven. Oxytocin is started at a rate of 1–2 milliunits/min and increased by 2 milliunits/min every 15 minutes to achieve a regular contraction pattern to a maximum infusion rate of 20 milliunits/min. Oxytocin is stopped for signs of fetal compromise, as defined by NICHD criteria or tachysystole. Before a revision to the protocol in 2015, providers could override the protocol to administer infusion rates up to 30 milliunits/min, though this was not standard. If oxytocin infusion is stopped, it cannot be restarted for 15 minutes. If it is restarted after less than 40 minutes and fetal status is reassuring, it can be administered at a rate one half of the rate that necessitated its discontinuation. If the infusion was discontinued for more than 40 minutes, the oxytocin is restarted at a dose of 1 or 2 milliunits/min and titrated as above.
Among nulliparous women who underwent induction of labor at term, patients who had at least 8 hours of continuous intravenous oxytocin infusion before entering the active phase of labor (defined here as protracted latent labor) were included in the analysis. Among those women, oxytocin rests that occurred before documentation of a dilation of 6 centimeters (cm) or greater were identified and calculated as the time from the discontinuation of oxytocin (rate of 0 milliunits/min) until the time when the infusion was restarted, as documented by the patient's nurse in the labor flowsheet. The longest oxytocin rest for each patient was used for the analysis. To confirm that these calculations correctly identified episodes and duration of oxytocin rest, 11 patients were identified at random and their medical record reviewed in detail, which validated the statistical and mathematical coding using the raw data from the EMR.
For the purposes of this analysis, the duration of each patient's latent phase was calculated as the time between initiation of oxytocin infusion and when a cervical dilation of 6 cm or greater was first documented in the labor flowsheet. For patients who underwent a cesarean delivery before achieving cervical dilation of 6 cm, the duration of the latent phase was calculated as the time when oxytocin infusion was first initiated until the delivery time of the neonate through cesarean. Among patients who underwent an oxytocin rest, the duration of the latent phase was adjusted by subtracting the duration of the oxytocin rest. Duration of the latent phase was found to have a nonnormal distribution. Therefore, for the purposes of the multivariable regression analysis, the latent phase was converted to a logarithmic scale.
As the duration of oxytocin rest increased, the number of patients exposed to each duration decreased. Among patients who had an oxytocin rest for more than 1 hour, a univariate analysis was performed to divide patients roughly equally into tertiles of 1 hour to less than 2 hours, 2 hours to less than 8 hours, and 8 hours or more.
The primary outcome of interest was mode of delivery (ie, cesarean vs vaginal delivery), which was extracted from the EMR. Secondary outcomes included maternal and neonatal morbidities. For maternal outcomes, we measured postpartum hemorrhage, chorioamnionitis, endometritis, venous thromboembolism, cerebrovascular accident, eclampsia, disseminated intravascular coagulation, and intensive care unit (ICU) admission, which were extracted from the EMR using ICD diagnosis codes and information from the EMR (Appendix 1, http://links.lww.com/AOG/B730).12,13 Internal hospital procedure codes were used to identify transfusion of red blood cells. A maternal composite outcome was counted as positive if the patient experienced any of the aforementioned diagnoses. Neonatal outcomes of interest, including meconium aspiration syndrome, moderate or severe infection, birth trauma, seizure, and hypoxic ischemic encephalopathy, were identified using relevant ICD diagnosis codes and information from the EMR.14 The need for respiratory support was identified by internal hospital procedure codes for respiratory therapy. A neonatal composite outcome was counted as positive if the neonate experienced any of the aforementioned diagnoses. Maternal and neonatal ICU admissions were also reported.
To determine whether additional cervical ripening methods were used during oxytocin rest, the charts of all patients who had an oxytocin rest of 4 hours or longer were individually reviewed to identify whether mechanical or pharmacologic cervical ripening was performed during the patient's longest oxytocin rest. At our institution, the protocol for induction of labor precludes administration of oxytocin for 4 hours after the most recent dose of prostaglandin, and, therefore, oxytocin rest of less than 4 hours was not considered long enough for additional cervical ripening to have occurred.
Chi-square test or Fisher exact test (for variables with expected cell counts less than 5) were employed to compare baseline characteristics among patients with continuous oxytocin, with brief (less than 1 hour) cessation of oxytocin, oxytocin rest lasting 1 to less than 2 hours, 2 hours to less than 8 hours, or 8 hours or longer. Analysis of variance and Kruskal-Wallis tests were used to compare normally and nonnormally distributed data. Characteristics that differed among the exposure groups were included in a multivariable logistic regression analysis to examine the association between oxytocin rest and mode of delivery. Based on the number of potential parameters, the initial model yielded a global shrinkage factor of 0.91, and a difference in apparent and adjusted Nagelkerke R2 of 0.01 (Appendix 2, available online at http://links.lww.com/AOG/B730).15 Using the partial F-statistic to evaluate the full and reduced models, we sequentially eliminated parameters that did not contribute meaningfully to the model to arrive at a model with optimized parsimony. Additionally, we used chi-square test or Fisher exact test (for variables with expected cell counts less than 5) to compare secondary outcomes (maternal and neonatal morbidities) among patients based on oxytocin rest duration. A multivariable regression analysis was performed to evaluate the association between neonatal intensive care unit (NICU) admissions with oxytocin rest while adjusting for maternal diabetes. All data analysis was performed using SAS 9.4. Statistical significance was determined based on two-sided tests with P value <.05.
A cohort of 3,136 nulliparous, term, vertex-presenting, singleton patients underwent induction of labor during the study period (Fig. 1). Of those, 1,193 patients remained in the latent phase after 8 hours of continuous oxytocin, and 267 patients (22.4%) underwent an oxytocin rest that lasted at least 1 hour. Our sample was 58.6% white, 18.4% black, 12.7% Hispanic, 6.0% Asian, and 4.4% listed as “other.”
Six hundred forty-two patients (53.8%) experienced no interruption of oxytocin infusion, 284 (23.8%) had oxytocin discontinuation for less than 1 hour, 106 (8.9%) had oxytocin rest 1 hour to less than 2 hours, 89 (7.4%) had oxytocin rest lasting 2 hours to less than 8 hours, and 72 (6.0%) had oxytocin rest lasting 8 hours or longer. Certain baseline characteristics differed significantly across the exposure groups (Table 1). Longer durations of oxytocin rest were associated with higher BMI, longer latent phase, hypertension, and diabetes (P<.05).
The overall rate of cesarean delivery in our cohort of nulliparous inductions of labor with at least 8 hours of oxytocin before entering active phase was 47.0% (n=561). A multivariable regression analysis was used to assess the association between duration of oxytocin rest and mode of delivery (Table 2). Significant confounding factors that contributed to our model included maternal age in years (adjusted odds ratio [aOR] 1.05 [95% CI 1.03–1.07]); gestational age in weeks (aOR 1.31 [95% CI 1.19–1.44], BMI (aOR 1.04 [95% CI 1.02–1.06]); the duration of latent phase in log (hours) (aOR 3.33 [95% CI 2.40–4.62]); diabetes (aOR 1.47 [95% CI 0.97–2.23]); and NICHD fetal heart rate category at the time of oxytocin rest (aOR 1.43 [95% CI 1.00–2.04]). Compared with parturients who received continuous oxytocin without rest, the risk of cesarean delivery decreased over the time intervals (P value for trend <.02); at less than 1 hour aOR 1.12 (95% CI 0.79–1.58), 1 hour to less than 2 hours aOR 0.78 (95% CI 0.48–1.27), 2 hours to less than 8 hours aOR 0.60 (95% CI 0.35–1.04); 8 hours or longer aOR 0.43 (95% CI 0.24–0.79). Patients in the cohort who had an oxytocin rest 8 hours or longer had a reduced risk of cesarean delivery compared with parturients who had no oxytocin rest.
To evaluate whether the association between oxytocin rest and mode of delivery was a result of additional cervical ripening that occurred during that interval, a multivariable regression analysis was performed to adjust for this factor. Sixty-five patients had additional cervical ripening, either misoprostol (35.9%), intrauterine balloon (54.7%) or both (9.4%) during oxytocin rest. In our multivariable regression analysis, additional cervical ripening during oxytocin rest was not associated with the odds of cesarean delivery, aOR 1.22 (95% CI 0.47–3.16) and did not affect the association between oxytocin rest and mode of delivery and was therefore not included in the final model.
To investigate whether the lower odds of cesarean delivery associated with oxytocin rest came at the expense of higher risk for maternal and neonatal morbidities, secondary outcomes were compared among patients with different oxytocin rest durations (Table 3). We did not detect a higher odds ratio (OR) of a composite of maternal or neonatal morbidity among women with an oxytocin rest in hours: OR 1.02 (95% CI 0.98–1.06) and OR 1.03 (95% CI 0.97–1.08), respectively. A potentially higher odds of NICU admission was noted with increasing duration of oxytocin rest, OR 1.04 (95% CI 1.00–1.08). However, after adjusting for the effect of maternal diabetes on NICU admission (aOR 3.73 [95% CI 2.48–5.62]), the odds of NICU admission fell, aOR 1.03 [95% CI 0.99–1.07], suggesting that NICU admissions may be more strongly influenced by other factors such as the institutional hypoglycemia protocol.
Oxytocin rest is a clinical practice sometimes used during inductions of labor when women remain in the latent phase, despite prolonged oxytocin exposure. There is biological plausibility borne out in the basic science literature that uterine myometrial cell response to oxytocin will diminish with prolonged continuous oxytocin exposure.4,7–10,16 In our cohort of nulliparous women undergoing induction of labor with a protracted latent phase, an oxytocin rest of at least 8 hours was associated with lower odds of cesarean delivery. Patients in this category had a higher BMI and were more likely to have hypertension and diabetes, yet this reduction in risk of cesarean delivery was not accompanied by significantly higher odds of maternal or neonatal morbidity. Importantly, our study was not powered to evaluate these uncommon secondary outcomes.
There are several mechanisms by which oxytocin rest might be protective against cesarean delivery. Allowing time for the oxytocin receptor to be upregulated and replenished on the myocyte cell surface is one possible mechanism. Oxytocin is also thought to act as a transcription factor to increase the expression of gap junction proteins, which synchronize contractions, and to increase the expression of prostaglandin F2a.17 Myometrial cells lose optimal functionality in a state of starvation ketosis; oxytocin rest may therefore represent an opportunity for patients to eat after a prolonged fasting period. It may also give patients an opportunity to ambulate, shower, and feel increased control over their induction course, which may have circulatory and psychological benefits.
Other factors may contribute to the apparent effect of an oxytocin rest. Because fetal heart rate abnormalities may be a reason for discontinuing oxytocin and may also be an indication for cesarean delivery, it is important to include the contribution of NICHD fetal heart rate category at the time of oxytocin rest in estimating the association between oxytocin rest and cesarean delivery. After adjusting for NICHD fetal heart rate category at the time of oxytocin rest, there was no association between oxytocin of rest less than 1 hour and cesarean delivery (Table 2). During oxytocin rest, providers may perform additional cervical ripening, but we did not find an association between cervical ripening and mode of delivery in our cohort.
One limitation of our study is that information about certain aspects of patient care were not available for analysis. First, whether oxytocin cessation was intended as an “oxytocin rest” by the provider is not available. Second, the timing of artificial rupture of membranes was not accounted for in our analysis for lack of consistent documentation in the EMR. However, it is not our practice to perform an intentional oxytocin rest in the setting of ruptured membranes. The interplay between these potential confounders and the effect of an oxytocin rest is less clear and is an important area of future investigation. Lastly, it should be noted that, although we included patients who had been on oxytocin without entering the active phase for 8 hours, this is not the definition of a protracted latent phase.
A potential source of bias in the study is that providers who employ oxytocin rest may also be more cautious about proceeding toward cesarean delivery and may be more likely to adhere to recommendations that prevent cesarean delivery.18 These unmeasured provider characteristics may confound our findings and warrant further investigation. Although this study demonstrates an association between oxytocin rest and mode of delivery, prospective, randomized studies should be undertaken to identify whether a causal relationship exists.
The rate of induction of labor is increasing and is performed for a quarter of U.S. births,1 a number that may increase further given growing evidence for reduced risk of cesarean delivery and gestational hypertension with induction of labor at 39 weeks of gestation in studies such as the ARRIVE trial.1,19–21 However, a failed induction of labor results in cesarean delivery,18 the prevention of which is an important goal in the field of obstetrics.18,22 Sometimes, despite cervical ripening and oxytocin infusion, patients remain in latent labor. When it is medically safe to do so, our results suggest that oxytocin rest for more than 8 hours may optimize a woman's chance of vaginal delivery.
1. Martin JA, Hamilton BE, Osterman MJK, Driscoll AK, Drake P. Births: final data for 2016. Natl Vital Stat Rep 2018;67:1–55.
2. Induction of labor. ACOG Practice Bulletin No. 107. American College of Obstetricians and Gynecologists. Obstet Gynecol 2009;114:386–97.
3. Phaneuf S, Asboth G, Carrasco MP, Linares BR, Kimura T, Harris A, et al. Desensitization of oxytocin receptors in human myometrium. Hum Reprod Update 1998;4:625–33.
4. Smith MP, Ayad VJ, Mundell SJ, McArdle CA, Kelly E, Lopez Bernal A. Internalization and desensitization of the oxytocin receptor is inhibited by Dynamin and clathrin mutants in human embryonic kidney 293 cells. Mol Endocrinol 2006;20:379–88.
5. Phaneuf S, Asboth G, Carrasco MP, Europe-Finner GN, Saji F, Kimura T, et al. The desensitization of oxytocin receptors in human myometrial cells is accompanied by down-regulation of oxytocin receptor messenger RNA. J Endocrinol 1997;154:7–18.
6. Phaneuf S, Rodriguez Linares B, TambyRaja RL, MacKenzie IZ, Lopez Bernal A. Loss of myometrial oxytocin receptors during oxytocin-induced and oxytocin-augmented labour. J Reprod Fertil 2000;120:91–7.
7. Crall HD, Mattison DR. Oxytocin pharmacodynamics: effect of long infusions on uterine activity. Gynecol Obstet Invest 1991;31:17–22.
8. Phaneuf S, Asboth G, MacKenzie IZ, Melin P, Lopez Bernal A. Effect of oxytocin antagonists on the activation of human myometrium in vitro: atosiban prevents oxytocin-induced desensitization. Am J Obstet Gynecol 1994;171:1627–34.
9. Robinson C, Schumann R, Zhang P, Young RC. Oxytocin-induced desensitization of the oxytocin receptor. Am J Obstet Gynecol 2003;188:497–502.
10. Balki M, Erik-Soussi M, Kingdom J, Carvalho JC. Oxytocin pretreatment attenuates oxytocin-induced contractions in human myometrium in vitro. Anesthesiology 2013;119:552–61.
11. Boie S, Glavind J, Velu AV, Mol BWJ, Uldbjerg N, de Graaf I, et al. Discontinuation of intravenous oxytocin in the active phase of induced labour. The Cochrane Database Systematic Review 2018, Issue 8. Art. No.: CD012274. DOI: 10.1002/14651858.CD012274.pub2.
12. Dos Santos F, Drymiotou S, Antequera Martin A, Mol BW, Gale C, Devane D, et al. Development of a core outcome set for trials on induction of labour: an international multistakeholder Delphi study. BJOG 2018;125:1673–80.
13. How does CDC identify severe maternal morbidity? Available at: https://www.cdc.gov/reproductivehealth/maternalinfanthealth/smm/severe-morbidity-ICD.htm
. Retrieved October 22, 2019.
14. Specifications manual for Joint Commission national quality measures. Version 2018B1. Available at: https://manual.jointcommission.org/releases/TJC2018B1/AppendixATJC.html
. Retrieved October 22, 2019.
15. Riley RD, Snell KI, Ensor J, Burke DL, Harrell FE Jr, Moons KG, et al. Minimum sample size for developing a multivariable prediction model: PART II—binary and time-to-event outcomes. Stat Med 2019;38:1276–96.
16. Balki M, Ramachandran N, Lee S, Talati C. The recovery time of myometrial responsiveness after oxytocin-induced desensitization in human myometrium in vitro. Anesth Analg 2016;122:1508–15.
17. Arrowsmith S, Wray S. Oxytocin: its mechanism of action and receptor signalling in the myometrium. J Neuroendocrinol 2014;26:356–69.
18. Safe prevention of the primary cesarean delivery. Obstetric Care Consensus No. 1. American College of Obstetricians and Gynecologists. Obstet Gynecol 2014;123:693–711.
19. Grobman WA, Bailit J, Lai Y, Reddy UM, Wapner RJ, Varner MW, et al. Defining failed induction of labor. Am J Obstet Gynecol 2018;218:122 e1–e8.
20. Grobman WA, Rice MM, Reddy UM, Tita ATN, Silver RM, Mallett G, et al. Labor induction versus expectant management in low-risk nulliparous women. N Engl J Med 2018;379:513–23.
21. Society of Maternal-Fetal Publications Committee. SMFM statement on elective induction of labor in low-risk nulliparous women at term: the ARRIVE trial. Am J Obstet Gynecol 2019;221:B2–4.
22. Banos N, Migliorelli F, Posadas E, Ferreri J, Palacio M. Definition of failed induction of labor and its predictive factors: two unsolved issues of an everyday clinical situation. Fetal Diagn Ther 2015;38:161–9.