Uterine atony is the most common cause of postpartum hemorrhage (PPH), and PPH is the leading cause of maternal mortality worldwide.1 Active management of the third stage of labor, which includes the prophylactic administration of oxytocin, has been shown to decrease uterine atony and the risk of PPH.2
Several studies have sought to determine the dose-response of IV oxytocin administered for third-stage prophylaxis. In patients undergoing scheduled cesarean delivery, the estimated effective dose in 90% of parturients (ED90) for satisfactory uterine tone of a bolus dose of oxytocin was 0.35 IU.3 In contrast, the ED90 in women undergoing cesarean delivery after intrapartum labor augmentation and arrest of labor was 2.99 IU.4 This difference in the oxytocin dose in laboring compared with nonlaboring women likely represents desensitization of the oxytocin receptor after prolonged oxytocin exposure.5 The density of oxytocin receptors, as well as the oxytocin mRNA concentration, was markedly lower in myocytes harvested at cesarean delivery from women with prolonged exposure to oxytocin during labor compared with women with no labor and prior oxytocin exposure.6 Given that many laboring women, especially those with arrest of labor, are exposed to exogenous oxytocin during labor, it is probable that the ED90 of third-stage oxytocin is higher in this patient population than in patients who undergo cesarean delivery without labor and prior exogenous oxytocin exposure.
When given as a rapid IV bolus, oxytocin has been associated with adverse reactions, such as hypotension, myocardial ischemia, and hemodynamic collapse.7,8 In an effort to mitigate hemodynamic changes after the administration of an IV bolus of oxytocin, infusions are often used clinically. Various studies have demonstrated the efficacy and safety of oxytocin infusions.8–10 George et al.11 estimated that the ED90 of an oxytocin infusion in women undergoing scheduled cesarean delivery is 0.29 IU/min. However, to our knowledge, there are no studies that have investigated the ED90 of oxytocin delivered via infusion to prevent uterine atony after cesarean delivery for dystocia and labor arrest. The aim of the current up-down sequential allocation dose-response study was to test the hypothesis that laboring parturients who receive intrapartum exogenous oxytocin therapy, and who subsequently undergo cesarean delivery for labor dystocia, will have a higher ED90 for oxytocin infusion in the third stage of labor compared with nonlaboring parturients.
This study was approved by the Northwestern University IRB (STU00059346), and written informed consent was obtained from all the study participants. The study design was a single-blinded, dual-arm, dose-response study using a 9:1 biased-coin sequential allocation method. Inclusion criteria were healthy parturients, 18 years or older, term gestation (≥37 weeks), ASA physical status class I or II, with a body mass index <40 kg/m2 undergoing cesarean delivery with neuraxial anesthesia. Exclusion criteria included general anesthesia, emergency cesarean delivery or nonreassuring fetal status, known placenta or uterine abnormalities, and patients receiving a magnesium sulfate infusion.
The experimental group consisted of laboring patients requiring cesarean delivery due to labor dystocia, defined as arrest of dilation in first stage of labor or arrest of descent in second stage of labor. Parturients in this group had labor induced or augmented with oxytocin and had neuraxial labor analgesia. Subjects were approached for study participation after the obstetrician’s decision to proceed to cesarean delivery. Epidural anesthesia was induced using 15 to 20 mL of 2% lidocaine with epinephrine 1:200,000 (±1 mL of sodium bicarbonate per 10 mL of lidocaine) administered in 5-mL aliquots to attain a T4 level of surgical anesthesia. Intrapartum exogenous oxytocin administration was discontinued at the time of the decision to perform cesarean delivery.
The control group consisted of nonlaboring women scheduled for cesarean delivery. Eligible women admitted to the Labor and Delivery Unit of Prentice Women’s Hospital were approached for study participation immediately after the routine preanesthetic evaluation. Spinal anesthesia was induced with hyperbaric bupivacaine 12 mg, fentanyl 15 μg, and morphine 150 μg. Subjects in both groups were convenience samples who were approached for study participation when the lead investigator (AL) was available. The investigator administered the oxytocin but did not otherwise participate in the anesthesia care of the subjects.
Subjects in both groups were positioned supine with left lateral tilt. Pulmonary aspiration prophylaxis consisting of IV ranitidine 150 mg, metoclopramide 10 mg, and oral citric acid/sodium citrate solution 30 mL was administered before transport to the operating room. Standard monitors, including pulse oximetry, noninvasive blood pressure cuff cycling every 2.5 minutes, 5-lead electrocardiogram, and side-stream capnography via nasal cannula, were applied upon arrival to the operating room. Preincision antibiotics were routinely administered. Vasopressors and IV fluids were administered at the anesthesiologist’s discretion per usual practice.
The study oxytocin infusion was initiated immediately after the umbilical cord was clamped. The starting infusion dose was 18 IU/h for the first patient in each group (approximately equates to 0.29 IU/min).11 The infusion was set up and administered by the principal investigator (AL). The patient, anesthesiologist, and surgeon were blinded to the dose. Uterine tone was assessed by the obstetrician 4 minutes after the initiation of the oxytocin infusion. Two outcomes were possible: satisfactory or unsatisfactory uterine tone. If the uterine tone was deemed unsatisfactory, the case was categorized as a failure and the infusion rate was increased to 36 IU/h. If the starting infusion rate was greater than 36 IU/h, the infusion rate was maintained. If uterine tone did not improve after an additional 3- to 6-minute period, an additional uterotonic agent (methylergonovine, 15-methyl prostaglandin F2α, or misoprostol) was administered. After a failed case, the next subject received an oxytocin infusion rate that was increased by 2 IU/h.
If uterine tone was deemed satisfactory, the case was categorized as a success and the next subject’s infusion rate was determined via a biased allocation method with a 90% chance to maintain the same infusion rate and a 10% chance to decrease the infusion rate by 2 IU/h. Biased allocation for each group was determined by drawing black and white marbles (ratio 1:9) out of an opaque bag. The allocation assignments were concealed in sequentially numbered opaque envelopes by an investigator not involved in the clinical care of the patient. The oxytocin infusion was interrupted at the end of the surgical procedure and then resumed in the postanesthesia care unit at a rate of 3.6 IU/h until discharge from the unit.
Subject characteristics, antepartum hemoglobin value, indication for cesarean delivery, total intraoperative dose of oxytocin, fetal weight, presence of polyhydramnios, macrosomia, and prior history of uterine atony were recorded. In the experimental group, the intrapartum oxytocin dose and duration, spontaneous or induced labor, and presence of chorioamnionitis at the time of cesarean delivery were also recorded. In addition to obstetrician-rated success or failure to achieve satisfactory uterine tone, the number of additional uterotonic agents, estimated blood loss, hemoglobin value on postoperative day 1, oxytocin-related adverse effects (headache, flushing, nausea, vomiting, and ST depression ≥0.2 mm), and the lowest recorded blood pressure in the 30-minute interval after initiation of the study drug were evaluated.
The sample size for this study was estimated by simulation assuming that the actual response would be described by a 2-parameter log-logistic function. Based on the starting infusion rate of 18 IU/h and an incremental dose of 2 IU/h, 100,000 simulations of 100 sequential cases were performed with estimated ED90 infusion rates between 20 and 36 IU/h for the nonlaboring group and between 36 and 46 IU/h for the laboring group. The a priori estimated probability of success for each case was 0.5, with a 1:9 biased allocation for the subsequent case after successful attainment of uterine tone. The increase in probability of success with increasing dose was simulated between 0.05 and 0.15. The stopping criterion was set at 90% successful cases of all cases at that dose. An upper limit of 100 subjects was set to avoid oversampling if the boundary could not be met.12 Based on the simulations, a minimal number of 20 subjects and a median between 24 and 40 subjects in each group was required to reach the end point in 90% of the simulations.
The primary outcome was the ED90 for satisfactory uterine tone as judged by the obstetrician. The study did not control for the obstetrician. Dose-response data for each group were evaluated by fitting a 2-parameter log-logistic function given by the expression f(x) = 1/(1 + exp(b(log(x) − log (e)))) and the ED90 estimates from the fitted equations using the delta method. The log-logistic model was chosen because oxytocin myometrium contractile effects have been shown to be symmetric around the ED50 point.5 Confidence intervals (CIs) of the differences in ED90 estimates were calculated using a bootstrap method with 9999 replications. Analysis of dose-response curves was performed using RKWard version 0.6.1, the package “drc” version 2.3-96, and R version 3.1.0 release date 4/10/2014 (The R Foundation for Statistical Computing, Vienna, Austria).13 Subject characteristics, the number of additional uterotonic agents required, estimated blood loss, hemoglobin value on postoperative day 1, and oxytocin-related adverse effects (headache, flushing, nausea, vomiting, and ST depression ≥ 0.2 mm) were compared between groups using the Mann-Whitney U test (interval or continuous data) or the Fisher exact test (binomial data). Median differences and CIs were calculated using a bootstrap method with 10,000 replications. The number of subjects receiving supplemental uterotonics and the 95% CI of the difference in proportions were compared using the 2-sample test for equality of proportions with continuity correction. The Pearson correlation coefficient was calculated to assess the relationship between the oxytocin rest interval (interval from discontinuation of labor oxytocin infusion to start of intraoperative infusion) and the estimated blood loss. The difference in the rest interval was compared in women who received and did not receive additional uterotonic agents. A P < 0.05 was required to reject the null hypothesis.
Between August 2012 and June 2013, 80 patients were approached and 9 declined participation. Seventy-one patients were included in the study, 38 in the nonlaboring group and 33 in the laboring group. Oxytocin was not delivered per study protocol due to a problem with the medication pump in 1 patient in the laboring group, and this patient was excluded from the analysis. Characteristics of the study groups are shown in Table 1. The percentage of nulliparous, gestational age, and the incidence of chorioamnionitis were greater in the laboring than in the nonlaboring group. Forty-nine obstetricians participated in the study, 30 in the nonlaboring group and 26 in the laboring group. Seven obstetricians provided care for subjects in both groups. Twenty-eight obstetricians managed 1, 15 managed 2, 2 managed 3, and 1 managed 6 subjects. The maximal number of subjects managed by a single obstetrician in either group was 4.
Labor was induced in 44% of the subjects in the laboring group. The median (interquartile range [IQR]) duration of oxytocin exposure in labor was 14 hours (11–19 hours), and the cumulative labor dose of oxytocin was 3.9 IU (1.8–6.5 IU). The oxytocin infusion was discontinued at the time of decision to proceed to cesarean delivery in all subjects in the laboring group. The median (IQR) time interval between the discontinuation of the labor oxytocin infusion and the study oxytocin infusion was 48 minutes (35–64 minutes). There was no association between the oxytocin rest interval and the estimated blood loss (Pearson r = −0.033; 95% CI, −0.244 to 0.351; P = 0.86). The median (IQR) oxytocin rest interval was 48 minutes (34–72 minutes) and 48 minutes (35–58 minutes) in subjects who did and did not receive additional uterotonic agents, respectfully (Mann-Whitney U test, P = 0.89).
Figure 1 shows the sequence of successful and failed oxytocin infusions during the third stage of labor for nonlaboring and laboring study subjects. The ED90 was significantly greater for the laboring group (44.2 IU/h [95% CI, 33.8–55.6]) compared with the nonlaboring group (16.2 IU/h [95% CI, 13.1–19.3]; difference in rate 28 IU/h [95% CI of difference, 26–29 IU/h]; P < 0.001).
Secondary outcomes are shown in Table 2. Thirty-four percent of patients in the laboring group required supplemental uterotonics compared with 8% in the nonlaboring group (difference 26% [95% CI of the difference, 7%–44%, P = 0.008]). Eight subjects (25%) in the laboring group received ≥2 uterotonic agents and 1 subject (3%) in the nonlaboring group received ≥2 uterotonic agents (difference 22% [95% CI of the difference, 4%–41%, P = 0.009]). The number of subjects experiencing an adverse effect was greater in the laboring group (difference 25% [95% CI of the difference, 10%–59%, P = 0.004]). Nausea and/or vomiting was the most prevalent adverse effect (difference 29% [95% CI of the difference, 4%–54%, P = 0.02]). Estimated blood loss was greater, and the hemoglobin value on the first postpartum day was lower in the laboring group than in the nonlaboring group.
The primary finding of this up-down sequential allocation study was that the ED90 of the oxytocin infusion rate administered for third stage of labor prophylaxis after cesarean delivery was almost 3 times greater in laboring women with prior exogenous oxytocin exposure than in women undergoing scheduled cesarean delivery without prior labor. In addition, more women with prior oxytocin exposure required an additional uterotonic agent to treat uterine atony compared with women without prior labor. The estimated ED90 in the control group (0.27 IU/min) was similar to that determined by George et al.11 (0.29 IU/min) in a similar group of women scheduled for cesarean delivery without labor.
There has been considerable discussion over the past 5 years surrounding the optimal dose and mode of administration for third-stage oxytocin administration.14–16 Although consensus is lacking regarding the optimal oxytocin dose, there appears to be general agreement that administration of a 10-IU IV bolus dose is a dangerous practice.14,16 Thus, many practitioners prefer to administer oxytocin via infusion rather than as a bolus.
Using an up-down sequential allocation method, George et al.11 estimated the ED90 for an oxytocin infusion in women undergoing elective cesarean delivery. However, there is evidence that higher oxytocin doses are required in women who have prior intrapartum exposure to exogenous oxytocin. In a 2004 study, Carvalho et al.3 estimated the ED90 in women scheduled for cesarean delivery without labor. In 2006, the same group of investigators used identical methods to estimate the ED90 of bolus-dose oxytocin in women undergoing cesarean delivery for the indication of labor dystocia.4 The ED90 in this latter group was 9-fold higher than that in the nonlaboring group. The higher ED90 infusion rate observed in the laboring group compared with nonlaboring women in the current study mimics the results of studies for bolus-dose oxytocin.
Oxytocin receptor desensitization in patients receiving large and prolonged doses of oxytocin during labor is well described and may lead to labor dystocia and increased risk for cesarean delivery, as well as risk for uterine atony, resulting in PPH.6,17–19
Phaneuf et al.6 demonstrated reduction in oxytocin binding sites in parturients who had induction of labor with oxytocin, suggesting downregulation of oxytocin receptors with prolonged exposure to oxytocin. Grotegut et al.18 conducted a retrospective study of women with severe PPH and found that these women received greater amounts of oxytocin during labor compared with women without PPH, suggesting that prolonged exposure to oxytocin leads to decreased drug efficacy, which may contribute to an increased risk of PPH. Similarly, in the current study, women in the laboring group with prior exposure to oxytocin had greater estimated blood loss and lower postoperative hemoglobin levels than the women not exposed to oxytocin, despite receiving higher doses of oxytocin. Current professional guidelines for third-stage oxytocin administration do not differentiate among women with prior oxytocin exposure and those without.20,21 Although oxytocin is considered the first-line drug for preventing and treating uterine atony, early consideration of additional uterotonic agents may be indicated in women with prior oxytocin exposure and oxytocin receptor desensitization and downregulation. Future studies should evaluate whether multimodal prophylactic uterotonic administration is more effective than oxytocin therapy alone.
Oxytocin is associated with major cardiovascular side effects.14 These include hypotension, myocardial ischemia, and arrhythmias. Nausea, vomiting, chest pain, headache, and flushing are less serious, but quite bothersome side effects that appear to be dose-dependent. For example, Jonsson et al.22 observed a higher incidence of ST-segment depression and chest pain in women randomly assigned to receive oxytocin 10 IU bolus compared with 5 IU after cesarean delivery. In our current study, the number of women with ST-segment changes was greater in the laboring than in nonlaboring group; however, the study was underpowered to identify a statistically significant difference between groups. One patient receiving 38 IU/h of oxytocin reported chest pain and had significant ST-segment depression and tachycardia in the setting of PPH. Overall, we also found a significant increase in side effects in the laboring group that received higher doses of oxytocin.
There are several limitations to our study design and conclusions. We used the biased-coin method for estimating the ED90 of the oxytocin infusion rate because it requires significantly fewer subjects than the traditional nonsequential dose-response design used to characterize the whole dose-response curve; we were only interested in comparing the ED90 between 2 groups. Simulation has demonstrated that the biased-coin modification of the traditional up-down sequential allocation method results in sufficient accuracy for estimation of the effective dose at high or low quantiles.12 However, the CIs of the point estimate may be wide unless the sample size is quite large. The sequential allocation method requires a binary outcome. We chose satisfactory uterine contraction as assessed by the obstetrician, as this was the same outcome used by George et al.11 However, this outcome is subjective and is not a validated outcome of the effect of oxytocin on the uterus. It is, however, used in daily clinical practice. We did not control for the participating obstetricians, and different obstetricians likely have different thresholds for “satisfactory” uterine tone. However, 49 different obstetricians participated in the study. In addition, although blinded to dose, the obstetricians were not blinded to patient group, and they may have had a different threshold for satisfactory uterine tone in women with prior labor compared with nonlaboring women. The laboring group included women with the diagnosis of chorioamnionitis. These women are at increased risk of uterine atony; thus, the estimated oxytocin ED90 might be lower in women without chorioamnionitis.
We chose to study the ED90 of an oxytocin infusion because the administration of oxytocin as an infusion, rather than a bolus, is common in the United States. However, from a pharmacokinetic standpoint, rapid attainment of a therapeutic plasma concentration of a drug is best achieved by a bolus dose followed by a maintenance infusion. Indeed, a bolus dose followed by an infusion was recommended by Stephens and Bruessel23 in their 2012 systematic review of oxytocin dosing at cesarean delivery. However, King et al.24 found no difference in the need for supplemental uterotonics in women at risk of uterine atony who were randomly assigned to receive a bolus followed by an infusion compared with a placebo-bolus followed by an infusion. Further study is required to address this issue.
Finally, our study evaluated early response to the oxytocin infusion. We did not assess the optimal duration of the oxytocin infusion, including whether the initial infusion rate can be decreased at a certain interval after delivery without increasing the risk of uterine atony.
In conclusion, the ED90 for oxytocin infusion during the third stage of labor is significantly higher in parturients who underwent cesarean delivery for dystocia and received exogenous intrapartum oxytocin compared with parturients undergoing elective cesarean delivery without labor and oxytocin exposure. These results are consistent with previous reports of oxytocin “resistance” in this population. We suggest that guidelines for oxytocin administration should differentiate between laboring patients with prior exposure to oxytocin and nonlaboring patients.
Name: Anne Lavoie, MD.
Contribution: This author helped design the study, conduct the study, and write the manuscript.
Conflicts: Anne Lavoie reported no conflicts of interest.
Attestation: This author has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: Robert J. McCarthy, DPharm.
Contribution: This author helped design the study, analyze the data, and write the manuscript.
Attestation: Robert J. McCarthy has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.
Name: Cynthia A. Wong, MD.
Contribution: This author helped design the study, conduct the study, and write the manuscript.
Attestation: Cynthia A. Wong has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
RECUSE NOTEDr. Cynthia A. Wong is the Section Editor for Obstetric Anesthesiology for the Journal. This manuscript was handled by Dr. Steven L. Shafer, Editor-in-Chief, and Dr. Wong was not involved in any way with the editorial process or decision.
1. Khan KS, Wojdyla D, Say L, Gülmezoglu AM, Van Look PF. WHO analysis of causes of maternal death: a systematic review. Lancet. 2006;367:1066–74
2. Begley CM, Gyte GM, Devane D, McGuire W, Weeks A. Active versus expectant management for women in the third stage of labour. Cochrane Database Syst Rev. 2011:CD007412
3. Carvalho JC, Balki M, Kingdom J, Windrim R. Oxytocin requirements at elective cesarean delivery: a dose-finding study. Obstet Gynecol. 2004;104:1005–10
4. Balki M, Ronayne M, Davies S, Fallah S, Kingdom J, Windrim R, Carvalho JC. Minimum oxytocin dose requirement after cesarean delivery for labor arrest. Obstet Gynecol. 2006;107:45–50
5. Kimura T, Saji F, Nishimori K, Ogita K, Nakamura H, Koyama M, Murata Y. Molecular regulation of the oxytocin receptor in peripheral organs. J Mol Endocrinol. 2003;30:109–15
6. Phaneuf S, Rodríguez Liñares B, TambyRaja RL, MacKenzie IZ, López Bernal A. Loss of myometrial oxytocin receptors during oxytocin-induced and oxytocin-augmented labour. J Reprod Fertil. 2000;120:91–7
7. Pinder AJ, Dresner M, Calow C, Shorten GD, O’Riordan J, Johnson R. Haemodynamic changes caused by oxytocin during caesarean section under spinal anaesthesia. Int J Obstet Anesth. 2002;11:156–9
8. McLeod G, Munishankar B, MacGregor H, Murphy DJ. Maternal haemodynamics at elective caesarean section: a randomised comparison of oxytocin 5-unit bolus and placebo infusion with oxytocin 5-unit bolus and 30-unit infusion. Int J Obstet Anesth. 2010;19:155–60
9. Munn MB, Owen J, Vincent R, Wakefield M, Chestnut DH, Hauth JC. Comparison of two oxytocin regimens to prevent uterine atony at cesarean delivery: a randomized controlled trial. Obstet Gynecol. 2001;98:386–90
10. Thomas JS, Koh SH, Cooper GM. Haemodynamic effects of oxytocin given as i.v. bolus or infusion on women undergoing Caesarean section. Br J Anaesth. 2007;98:116–9
11. George RB, McKeen D, Chaplin AC, McLeod L. Up-down determination of the ED(90) of oxytocin infusions for the prevention of postpartum uterine atony in parturients undergoing Cesarean delivery. Can J Anaesth. 2010;57:578–82
12. Pace NL, Stylianou MP. Advances in and limitations of up-and-down methodology: a précis of clinical use, study design, and dose estimation in anesthesia research. Anesthesiology. 2007;107:144–52
13. Knezevic SZ, Streibig JC, Ritz C. Utilizing R software package for dose-response studies. Weed Technol. 2007;21:840–8
14. Dyer RA, van Dyk D, Dresner A. The use of uterotonic drugs during caesarean section. Int J Obstet Anesth. 2010;19:313–9
15. Balki M, Tsen L. Oxytocin protocols for cesarean delivery. Int Anesthesiol Clin. 2014;52:48–66
16. Tsen LC, Balki M. Oxytocin protocols during cesarean delivery: time to acknowledge the risk/benefit ratio? Int J Obstet Anesth. 2010;19:243–5
17. Grotegut CA, Feng L, Mao L, Heine RP, Murtha AP, Rockman HA. β-Arrestin mediates oxytocin receptor signaling, which regulates uterine contractility and cellular migration. Am J Physiol Endocrinol Metab. 2011;300:E468–77
18. Grotegut CA, Paglia MJ, Johnson LN, Thames B, James AH. Oxytocin exposure during labor among women with postpartum hemorrhage secondary to uterine atony. Am J Obstet Gynecol. 2011;204:56.e1–6
19. Robinson C, Schumann R, Zhang P, Young RC. Oxytocin-induced desensitization of the oxytocin receptor. Am J Obstet Gynecol. 2003;188:497–502
20. American College of Obstetricians and Gynecologists. . ACOG Practice Bulletin: Clinical Management Guidelines for Obstetrician-Gynecologists Number 76, October 2006: postpartum hemorrhage. Obstet Gynecol. 2006;108:1039–47
21. Leduc D, Senikas V, Lalonde AB, Ballerman C, Biringer A, Delaney M, Duperron L, Girard I, Jones D, Lee LS, Shepherd D, Wilson KClinical Practice Obstetrics Committee; Society of Obstetricians and Gynaecologists of Canada. . Active management of the third stage of labour: prevention and treatment of postpartum hemorrhage. J Obstet Gynaecol Can. 2009;31:980–93
22. Jonsson M, Hanson U, Lidell C, Nordén-Lindeberg S. ST depression at caesarean section and the relation to oxytocin dose. A randomised controlled trial. BJOG. 2010;117:76–83
23. Stephens LC, Bruessel T. Systematic review of oxytocin dosing at caesarean section. Anaesth Intensive Care. 2012;40:247–52
24. King KJ, Douglas MJ, Unger W, Wong A, King RA. Five unit bolus oxytocin at cesarean delivery in women at risk of atony: a randomized, double-blind, controlled trial. Anesth Analg. 2010;111:1460–6