Abstract
Aim: We examined whether a reduction in the dose of uterotonic agents before amniotomy affects the amount of intrapartum hemorrhage.
Methods: Women who underwent planned painless vaginal delivery through labor induction with oxytocin and amniotomy at our hospital (n=294) were included in this study. The amount of intrapartum hemorrhage was compared among patients divided into the following two groups: (1) the oxytocin down dose (ODD) group (n=134), in which the dose of oxytocin was halved before amniotomy (from January to December 2023); and (2) the non-ODD group (n=160), in which the dose of oxytocin was not reduced before amniotomy (from January to December 2022).
Results: The incidence of postpartum massive hemorrhage (PPMH) did not differ between the two groups. However, while obstetrical disseminated intravascular coagulation (DIC) was not observed in the ODD group, five patients in the non-ODD group experienced obstetrical DIC, although not significant. Among patients with obstetrical DIC, uterine-type amniotic fluid embolism (AFE) was suspected in two patients.
Conclusions: The findings of this study indicate that, while amniotomy for labor induction poses a potential risk of obstetrical DIC and AFE, careful use of uterotonic agents may reduce this risk.
Introduction
According to a publication from the Ministry of Health, Labour and Welfare in 2020, painless delivery is offered in 26.0% of all delivery facilities in Japan, with only 8.6% of patients selecting this method for childbirth.1) There are two types of painless delivery: (1) 24-h painless delivery, which begins after the onset of spontaneous labor, and (2) planned painless delivery, which is performed using labor induction.
Labor induction is a risk factor for postpartum hemorrhage (PPH) of ≥500 g and amniotic fluid embolism (AFE).2,3) During painless delivery, amniotomy is performed in some cases because of slow labor progress, but it is not routinely recommended; according to a meta-analysis, amniotomy is associated with an increase in the rate of cesarean section, albeit not significantly.4) The WHO also recommends against customary amniotomy in the first stage of labor to prevent prolonged labor.5) Meanwhile, the rate of cesarean section was found to be lower with active delivery management (including amniotomy and labor augmentation with oxytocin) than without, and prolonged labor was also avoided.6) Therefore, in clinical practice, amniotomy is often performed in patients undergoing painless delivery.
However, amniotomy is associated with a risk of postpartum massive hemorrhage (PPMH).7) Oxytocin, a uterotonic agent, is frequently used to induce labor and stop bleeding in PPMH. An overdose of uterotonic agents and amniotomy can pose a risk of severe labor pains and PPMH.
Since January 2023, our hospital has implemented a reduction in the dose of uterotonic agents by half before amniotomy performed during labor induction in order to reduce the risk of PPMH. In this study, we retrospectively examined whether a reduction in uterotonic agent dose prior to amniotomy affected the amount of intrapartum hemorrhage.
Materials and methods
A total of 294 women who underwent planned painless vaginal delivery through labor induction with oxytocin and amniotomy at our hospital between January 2022 and December 2023 were included in this study. Patients were divided into the following two groups:
1. The oxytocin down dose (ODD) group (n=134): the oxytocin dose was halved just before amniotomy (from January to December 2023).
2. The non-ODD group (n=160): the oxytocin dose was not reduced before amniotomy (from January to December 2022).
The incidence rates of PPMH, critical obstetrical hemorrhage, and obstetrical disseminated intravascular coagulation (DIC) were compared between the two groups. Because a shock index ≥1.0 (vaginal delivery: blood loss ≥1,000 g) is used to define PPMH in the “Guidelines for Management of Critical Obstetrical Hemorrhage 2022”,8) we used intrapartum blood loss ≥1,000 g to define PPMH in this study. In the aforementioned guidelines, “critical obstetrical hemorrhage” is defined as PPMH occurring concomitantly with “persistent bleeding and abnormal vital signs,” “shock index ≥1.5,” “obstetrical DIC score ≥8,” or “fibrinogen level <150 mg/dl”. In this study, all mothers diagnosed with critical obstetrical hemorrhage were transferred to a high-tier medical care facility for multidisciplinary treatment such as blood transfusion. In such cases, the amount of blood loss was defined as the amount that could be measured at our hospital until the patient was transferred. In cases of critical obstetrical hemorrhage, a tentative version of the diagnostic criteria for obstetrical DIC9) was used to determine whether the patient had obstetrical DIC (Table 1). Obstetrical DIC could be diagnosed when the obstetrical DIC score reached ≥8 points (Table 1).
Table 1. Tentative version of diagnostic criteria for obstetrical disseminated intravascular coagulation
9)| Underlying disease/sign | Score | Laboratory findings
(coagulation system) | Score | Laboratory findings
(fibrinolytic system) | Score |
|---|
| | Fibrinogen (mg/dl) | | a. FDP (μg/ml) | |
| a. Abruption placenta | 4 | ≥300 | 0 | <30 | 0 |
| | 200–300 | 1 | 30–60 | 1 |
| b. Amniotic fluid embolism | 4 | 150–200 | 2 | ≥60 | 2 |
| | 100–150 | 3 | b. D-dimer (μg/ml) | |
| c. Noncoagulable postpartum hemorrhage† | 4 | <100 | 4 | <15 | 0 |
| | | 15–25 | 1 |
| | | | ≥25 | 2 |
Note: For puerperal women with abnormal postpartum hemorrhage that is difficult to stop, select and compare only one of the following items: underlying disease/sign, coagulation test, and fibrinolytic system test. Puerperal women with a score ≥8 are diagnosed with DIC.
† Noncoagulable postpartum hemorrhage refers to postpartum hemorrhage in which bleeding is not accompanied by a clot. It is desirable to collect blood in a container such as a pus basin to make sure that no clots form.
The hemoglobin/fibrinogen (H/F) ratio was used to determine whether the patient had coagulopathy.10) As an interpretation of this index, an H/F ratio ≥100 is considered to indicate a complication of DIC.10)
AFE was diagnosed using the diagnostic criteria for clinical AFE (Table 2)3) and serum markers. Zinc coproporphyrin (Zn-CP1), sialyl Tn antigen (STN), compliment components C3 and C4, and C1 esterase inhibitor (C1 inhibitor) are known as serum markers for AFE.11,12) Serum marker data were submitted at the discretion of the high-tier medical care facility.
Table 2. Clinical diagnostic criteria for amniotic fluid embolism
3)(1) Onset during pregnancy or within 12 h after delivery
(2) Intensive medical treatment for one or more of the symptoms or diseases listed under (A)–(D) below
(A) Cardiac arrest
(B) Respiratory failure
(C) Disseminated intravascular coagulation (DIC)
(D) Unexplained massive bleeding (>1,500 ml) more than 2 h after delivery
(3) If the observed findings and symptoms cannot be explained by any disease other than amniotic fluid embolism, a diagnosis of amniotic fluid embolism is clinically made when (1), (3), and at least one of (A)–(D) are met. |
Labor induction was performed using a uterotonic agent (oxytocin) for patients with a Bishop score ≥5 in principle according to the Guidelines for Obstetrical Practice in Japan 2020.13) The labor induction protocol was performed using oxytocin 5 IU with 5% purified glucose 500 ml. The starting dose was 2 mIU/min, which was increased by 2 mIU/min up to a maximum dose of 20 mIU/min until regular uterine contractions were achieved. After reducing the oxytocin dose prior to amniotomy, if contractions were weak, the oxytocin dose was increased according to the above protocol. Mechanical cervical ripening using a metreurynter or Laminaria tent was not performed. For patients with a Bishop score <5 points after 41 weeks of gestation, labor was induced while using a tocomonitor to carefully check for fetal heart rate findings indicative of hyperdynamia uteri and uterine rupture.
Amniotomy was performed in all patients. In cases where amniotomy was performed with oxytocin, it failed to induce cervical ripening or induced only weak labor if the fetal head position was at or below station −3 cm. Painless delivery was performed using epidural anesthesia, and anesthetics were used from the onset of cervical ripening.
Vacuum extraction was performed in cases where the fetal head position was at or below +3 station. No cases of retention of the placenta, total hysterectomy, sepsis, or maternal death were observed. One obstetrician-gynecologist determined Bishop scores and managed childbirth in all cases. Intrauterine balloon tamponade was performed in patients diagnosed with PPMH and atonic bleeding.
Statistical analysis was performed using the χ2 test and Mann–Whitney U test. Bell Curve for Excel (Social Survey Research Information Co., Ltd.) was used for the analysis. This study was approved by the Ethics Committee of Nakabayashi Hospital (Approval No. 23-N001), Tokyo, Japan, and was performed in compliance with the Declaration of Helsinki. Although consent was not obtained from patients, the presented data were anonymized, avoiding the risk of identification.
Results
There were no differences in age, maternal height, pregestational body mass index (BMI), primiparity rate, weeks of gestation at delivery, birth weight, umbilical cord arterial pH, and Bishop scores at labor induction and amniotomy between the ODD and non-ODD groups (Table 3). Vacuum extraction was performed significantly more frequently in the ODD group than in the non-ODD group (52.2% vs. 26.3%; p<0.01) (Table 4). In cases of vacuum extraction, the rate of weak pain as an indication for vacuum extraction was significantly higher in the ODD group than in the non-ODD group (28.6% vs. 11.9%; p<0.05) (Table 5).
Table 3. Background variables in the ODD and non-ODD groups
| ODD group (n=134) | Non-ODD group (n=160) | P |
|---|
| Age (years) | 33.0 (20–43) | 32.0 (22–45) | NS |
| Height (cm) | 160 (143–175) | 159 (145–171) | NS |
| Pregestational BMI (kg/m2) | 20.6 (16.3–27.6) | 20.9 (16.0–26.9) | NS |
| Primiparity rate (%) | 57.5 | 49.4 | NS |
| Weeks of gestation at delivery (weeks) | 38.0 (37–41) | 38.0 (37–41) | NS |
| Birth weight (g) | 2,941 (2,362–4,198) | 2,944 (2,354–3,812) | NS |
| Umbilical cord arterial pH | 7.33 (7.19–7.43) | 7.32 (7.10–7.44) | NS |
| Bishop score at induction of labor | 6.0 (3.0–9.0) | 5.0 (3.0–8.0) | NS |
| Bishop score at amniotomy | 7.0 (3.0–11.0) | 7.0 (4.0–11.0) | NS |
| Median (range) | | | |
ODD group, oxytocin down dose group; Non-ODD group, non-oxytocin down dose group; BMI, body mass index; NS, not significant
Table 4. Perinatal events in the ODD and non-ODD groups
| ODD group (n=134) | Non-ODD group (n=160) | P |
|---|
| Vacuum extraction (%) | 52.2 | 26.3 | <0.01 |
| Cervical laceration (%) | 7.5 | 18.1 | <0.01 |
| Intrapartum blood loss (g) | 338 (78.0–1,784) | 399 (69.0–2,540) | NS |
| PPMH | 10.4 | 10.6 | NS |
| PPMH and intrauterine tamponade (%) | 0.75 | 5.63 | <0.05 |
| PPMH and cervical laceration (%) | 0.75 | 5.63 | <0.05 |
| Critical obstetrical hemorrhage and obstetrical DIC (%) | 0 | 3.1 | NS |
| Median (range) | | | |
PPMH, postpartum massive hemorrhage; DIC, disseminated intravascular coagulation; NS, not significant
Table 5. Indications for vacuum extraction in the ODD and non-ODD groups
| ODD group (n=70) | Non-ODD group (n=42) | P |
|---|
| Weak pain (%) | 28.6 | 11.9 | <0.05 |
| Arrest of labor (%) | 18.6 | 21.4 | NS |
| Nonreassuring fetal status (%) | 51.4 | 64.3 | NS |
| Others (%) | 1.4 | 2.4 | NS |
NS, not significant
The median amount of intrapartum hemorrhage (338 vs. 399 g) and the incidence of PPMH (10.4% vs. 10.6%) did not differ significantly between the two groups (Table 4). However, cervical laceration was noted significantly less frequently among patients with PPMH in the ODD group than in the non-ODD group (0.8% vs. 5.6%; p<0.01). Significantly fewer patients underwent intrauterine balloon tamponade for PPMH in the ODD group than in the non-ODD group (0.8% vs. 5.6%; p<0.01) (Table 4).
Obstetrical DIC was not observed in the ODD group but was observed in five patients (3.1%) in the non-ODD group, although the difference between groups was not significant (Table 4). Background variables and blood test findings in cases of obstetrical DIC are shown in Table 6. In cases 1–3, fibrinogen levels were <150 mg/dl and H/F ratios were ≥100 (Table 6). Serum markers of AFE were also tested in cases 1 and 2, revealing low levels of C3 and C1 inhibitor in case 1 and high STN in case 2 (Table 7).
Table 6. Background variables and blood test findings in cases of obstetrical disseminated intravascular coagulation
| Case | Age
(years) | Primiparous/
multiparous | Weeks of gestation at delivery
(weeks) | Birth weight
(g) | Amount of blood loss
(g) | Blood tests before maternal transfer | Blood tests at hospitals accepting transfer |
|---|
Hemoglobin
(g/dl) | Fibrinogen
(mg/dl) | H/F ratio | Fibrinogen
(mg/dl) | FDP
(μg/ml) | D-dimer
(μg/ml) | Obstetrical DIC score |
|---|
| 1 | 27 | Primiparous | 37 | 2,686 | 1,666 | 12.8 | 114 | 112 | 69 | 811 | 393 | 10 |
| 2 | 39 | Multiparous | 38 | 2,768 | 2,320 | 11.1 | 80 | 139 | 32 | 1,200 | 600 | 10 |
| 3 | 31 | Primiparous | 40 | 3,158 | 2,180 | 9.8 | 80 | 123 | 50 | 413 | 180 | 10 |
| 4 | 36 | Multiparous | 38 | 2,940 | 936 | 11 | 160 | 69 | 115 | 377 | 103 | 9 |
| 5 | 32 | Multiparous | 37 | 3,438 | 2,540 | 9.7 | 206 | 47 | 189 | 92.4 | 33.3 | 8 |
H/F ratio, hemoglobin/fibrinogen ratio; FDP, fibrin degradation product
Table 7. Serum markers of amniotic fluid embolism in cases of obstetrical disseminated intravascular coagulation
| Case | Zn-CP1
(normal ≤1.6 pmol/ml) | STN
(normal ≤45 IU/ml) | C3
(normal 80–140 mg/dl) | C4
(normal 11–34 mg/dl) | C1 inhibitor
(normal >42%) |
|---|
| 1 | <1.6 | 19 | 77 (L) | 15 | <25 (L) |
| 2 | <1.6 | 210 (H) | 81 | 18 | 31 (L) |
Zn-CP1, Zinc coproporphyrin; STN, Sialyl Tn; C1 inhibitor, C1 esterase inhibitor
Discussion
In this study, a reduction in uterotonic agent dose prior to amniotomy did not affect the amount of intrapartum hemorrhage or the onset of PPMH. While amniotomy during labor induction for painless delivery was suggested to be a possible risk factor for critical obstetrical hemorrhage and obstetrical DIC, this risk may be minimized by reducing the dose of uterotonic agents before amniotomy.
The causes of PPMH include uterine atony, retention of the placenta, laceration of the birth canal, and coagulation disorders.14) In obstetrical DIC, dilutional coagulopathy is caused by massive hemorrhage, whereas consumption coagulopathy is caused by underlying diseases such as placental abruption and AFE. AFE, an underlying disorder of consumption coagulopathy, is classified into cardiopulmonary collapse-type AFE and uterine-type AFE; the former is primarily manifested as respiratory and circulatory failure and coagulopathy, and the latter as atonic hemorrhage and coagulopathy.3,15) In uterine-type AFE, increased uterine vascular permeability and severe uterine atony occur as anaphylactoid reactions to amniotic fluid that enters the maternal bloodstream.3) This is accompanied by the manifestation of hypercoagulable and hyperfibrinolytic states, leading to DIC and atonic hemorrhage.3) The diagnosis is made pathologically when uterine tissue is available through total hysterectomy or other means.11) However, the diagnostic criteria for clinical AFE are used in clinical practice because pathological diagnosis is time-consuming (Table 2).3) When pathological tissue cannot be obtained, serum markers are used for auxiliary diagnosis. Zn-CP1, STN, complement components C3 and C4, and C1 inhibitor are known serum markers for AFE. Because Zn-CP1 and STN are abundant in amniotic fluid and meconium, the detection of these markers in maternal blood is considered to indicate that fetal components have entered the maternal bloodstream.11) C3 and C4 are enzymes (complements) that complement the antigen–antibody reaction and have been reported to decrease in AFE due to anaphylactoid reactions.12) C1 inhibitor, a regulatory protein of the complement system, coagulation system, and kinin production system, has also been reported to decrease in AFE.12)
In this study, critical obstetrical hemorrhage and obstetrical DIC were observed only in the group of women who did not undergo uterotonic agent dose reduction. However, these findings were not significant owing to the rarity of critical events such as obstetrical DIC and AFE, as well as the small sample size. Of the five cases presenting with obstetrical DIC, cases 1–3 had consumption coagulopathy based on H/F ratios (Table 6) and also met the criteria for clinical AFE (Table 2). In cases 1 and 2, positive serum markers suggested that these patients likely had uterine-type AFE (Table 7).
AFE is a rare, severe maternal complication with an incidence of 0.001%–0.013%.16,17) The incidence of AFE at our hospital in 2022 was significantly higher, possibly because all patients with painless delivery underwent amniotomy with labor induction.
According to a review from Japan, both amniotomy and entry of amniotic fluid into the cervix are risk factors for AFE.11) Moreover, a study using a squamous cell carcinoma (SCC) antigen in maternal blood reported that, in mothers with cervical laceration, amniotic fluid components were found in maternal blood at levels higher than usual, although a small amount of amniotic fluid physiologically enters myometrial tissue during vaginal delivery.18) These findings suggest that amniotomy during the use of uterotonic agents may pose a potential risk of critical obstetrical hemorrhage and obstetrical DIC due to uterine-type AFE. As a possible mechanism, amniotomy during the use of uterotonic agents could lead to laceration and tissue contusions in the cervix because of increased uterine contractions and intrauterine pressure; this could cause amniotic fluid to enter the maternal bloodstream through blood vessels in the collapsed tissue, resulting in AFE and obstetrical DIC.
This study was conducted in women who underwent painless delivery. Because of workforce shortage and other issues, our hospital provides planned painless delivery under anesthesia, in which labor is induced during the daytime. Amniotomy is frequently used to induce labor or to expedite labor in cases of weak labor. The abovementioned mechanism leading to AFE and obstetric DIC can occur in both painless and spontaneous deliveries, although in painless delivery, amniotomy and excessive use of uterotonic agents may be more common, as anesthetics slow the progression of labor and delivery and reduce labor pain. As noted above, painless delivery is adopted at a low rate in Japan; however, with many pregnant women preferring painless delivery, the number of facilities offering painless delivery is expected to increase. Consequently, facilities that perform planned painless deliveries may face increased risks of critical obstetrical hemorrhage and obstetrical DIC.
The present findings suggest that a reduction in the dose of uterotonic agents prior to amniotomy, i.e., increasing awareness of hyperdynamia uteri, may reduce the risk of critical obstetrical hemorrhage and obstetrical DIC. Meanwhile, PPMH could not be reduced by these measures. This was likely due to the increased use of vacuum extraction as a result of the reduced dose of uterotonic agents, with bleeding attributed to laceration of the soft birth canal (i.e., the vaginal wall), rather than atonic hemorrhage, as indicated by the infrequent use of intrauterine balloon tamponade. A future study should focus on hemostatic and suturing methods for soft birth canal laceration to reduce postpartum blood loss.
The limitations of this study are as follows. The use of external tocometry prevented the evaluation of intrauterine pressure after amniotomy. In addition, since critical events such as obstetrical DIC and AFE are inherently rare, the sample size was small, being restricted to a single hospital in the department of obstetrics and gynecology.
Amniotomy for labor induction during planned painless delivery is a potential risk factor for critical obstetrical hemorrhage, obstetrical DIC, and AFE, but the findings of this study indicate that this risk can be reduced by careful use of uterotonic agents.
Acknowledgments
All authors have contributed significantly and agree with the content of the manuscript. The authors thank Ulatus (www. Ulatus. jp) for the English language review.
Conflict of interest
The authors have no conflict of interests to declare.
References
- 1. Ministry of Health, Labour and Welfare of Japan. Overview of medical facilities (static and dynamic) survey (confirmed Number) and hospital reports in 2022 (In Japanese). Available from URL: https://www.mhlw.go.jp/toukei/saikin/hw/iryosd/20/index.html. Accessed March 1, 2024.
- 2. Sheiner E, Sarid L, Levy A, Seidman DS, Hallak M. Obstetric risk factors and outcome of pregnancies complicated with early postpartum hemorrhage: a population-based study. J Matern Fetal Neonatal Med. 2005; 18: 149–154.
- 3. Kanayama N, Tamura N. Amniotic fluid embolism: pathophysiology and new strategies for management. J Obstet Gynaecol Res. 2014; 40: 1507–1517.
- 4. Smyth RMD, Alldred SK, Markham C. Amniotomy for shortening spontaneous labour. Cochrane Database Syst Rev. 2013; 6: CD006167.
- 5. World Health Organization. WHO recommendations: intrapartum care for a positive childbirth experience, 2018.
- 6. Brown HC, Paranjothy S, Dowsell T, et al. Package of care for active management in labour for reducing cesarean section rates in low-risk women. Cochrane Database Syst Rev. 2008: CD004907.
- 7. Nakabayashi Y, Santo M, Nakabayashi M, et al. Amniotomy with epidural analgesia in labor is the risk of postpartum massive hemorrhage. J Obstet Gynecol Neonat Hematol. 2020; 30: 60–68 (In Japanese).
- 8. Japan Society of Obstetrics and Gynecology. Japanese clinical practice guide for critical obstetrical hemorrhage (In Japanese). Available from URL: https://www.jsog.or.jp/activity/pdf/shusanki_taioushishin2022.pdf. Accessed March 1, 2024.
- 9. Itakura A, Serizawa M, Takeda J, et al. Disseminated intravascular coagulation (DIC) in obstetrics field: diagnostic criteria for obstetrical DIC (Tentative version). J Obstet Gynecol Neonat Hematol. 2023; 32: 43–49 (In Japanese).
- 10. Oda T, Tamura N, Ide R, et al. Consumptive coagulopathy involving amniotic fluid embolism: the importance of earlier assessments for interventions in critical care. Crit Care Med. 2020; 48: e1251–e1259.
- 11. Kanayama N. Amniotic fluid embolism: pathophysiology and management. J Obstet Gynecol Neonat Hematol. 2019; 28: 5–16 (In Japanese).
- 12. Tamura N, Kimura S, Farhana M, et al. C1 esterase inhibitor activity in amniotic fluid embolism. Crit Care Med. 2014; 42: 1392–1396.
- 13. Itakura A, Satoh S, Aoki S, et al. Guidelines for obstetrical practice in Japan: Japan Society of Obstetrics and Gynecology and Japan Association of Obstetricians and Gynecologists 2020 edition. J Obstet Gynaecol Res. 2023; 49: 5–53.
- 14. Mclintock C, James AH. Obstetric hemorrhage. J Thromb Haemost. 2011; 9: 1441–1451.
- 15. Kanayama N, Inori J, Ishibashi-Ueda H, et al. Maternal death analysis from the Japanese autopsy registry for recent 16 years: significance of amniotic fluid embolism. J Obstet Gynaecol Res. 2011; 37: 58–63.
- 16. Knight M, Berg C, Brocklehurst P, et al. Amniotic fluid embolism incidence, risk factor and outcomes: a review and recommendations. BMC Pregnancy Childbirth. 2012; 12: 7.
- 17. Stein PD, Matta F, Yaekoub AY. Incidence of amniotic fluid embolism: relation to cesarean section and to age. J Womens Health. 2009; 18: 327–329.
- 18. Koike N, Iwai K, Akasaka J. Entry of amniotic fluid into maternal circulation identified with SCC. J Obstet Gynecol Neonat Hematol. 2015; 25: S103–S104 (In Japanese).
