Volume 91, Issue 10 p. 1177-1183
MAIN RESEARCH ARTICLE
Free Access

Selective vaginal breech delivery at term – still an option

ELLI TOIVONEN

ELLI TOIVONEN

University of Tampere, Tampere

Search for more papers by this author
OUTI PALOMÄKI

OUTI PALOMÄKI

Department of Obstetrics and Gynecology, Tampere University Hospital, Tampere

Search for more papers by this author
HEINI HUHTALA

HEINI HUHTALA

University of Tampere, Tampere

Research Unit, Tampere University Hospital, Tampere, Finland

Search for more papers by this author
JUKKA UOTILA

JUKKA UOTILA

Department of Obstetrics and Gynecology, Tampere University Hospital, Tampere

Search for more papers by this author
First published: 18 June 2012
Citations: 47
E. Toivonen, University of Tampere, School of Medicine, 33041 Tampere, Finland. E-mail: [email protected]

Conflict of interest: The authors have stated explicitly that there are no conflicts of interest in connection with this article.

Please cite this article as: Toivonen E, Palomäki O, Huhtala H, Uotila J. Selective vaginal breech delivery at term – still an option. Acta Obstet Gynecol Scand. 2012; 91:10.1111/j.1600-0412.2012.01488.x.

Abstract

Objective. To compare the neonatal outcome between planned vaginal or planned cesarean section (CS) breech delivery and planned vaginal vertex delivery at term with singleton fetuses. Design. A cohort study. Setting. Delivery Unit, Tampere University Hospital, Finland, with 5200 annual deliveries. Population. The term breech deliveries over a period of five years (January 2004 to January 2009), a total of 751 breech deliveries, and 257 vertex controls. Methods. The data were collected from the mother's medical records, including a summary of the newborn. In the case of neonatal health problems, the pediatric records were also examined. Main outcome measures. Maternal and neonatal mortality and morbidity as defined in the Term Breech Trial. Low Apgar scores or umbilical cord pH as secondary end-points. Results. There was no neonatal mortality. Severe morbidity was rare in all groups, with no differences between groups. The Apgar scores at one minute were lower in the planned vaginal delivery group compared with the other groups, but there was no difference at the age of five minutes. Significantly more infants in the vaginal delivery group had a cord pH < 7.05. There was one maternal death due to a complicated CS in the planned CS group and none in the other groups. Mothers in the planned CS group suffered significantly more often from massive bleeding and needed transfusions. Conclusions. Vaginal delivery remains an acceptable option for breech delivery in selected cases.

Abbreviations:

  • CS
  • cesarean section
  • NICU
  • neonatal intensive care unit
  • planned CS
  • planned breech cesarean section
  • planned VD
  • planned vaginal breech delivery
  • Introduction

    About 3–4% of fetuses remain in breech presentation at term. A breech birth seems to predispose newborn children to asphyxia and birth traumas, and thus the method of delivery has been a subject of debate for decades. The Term Breech Trial by Hannah et al. in 2000 is by far the largest randomized controlled trial on the subject. The results of that study strongly favored elective cesarean section as the method of delivery; neonatal mortality and serious morbidity were threefold in the planned vaginal delivery group compared with the elective cesarean section (CS) group (1). The study affected clinical practice and guidelines in many countries (2–4), and subsequently, some studies showed reduced neonatal mortality and morbidity rates due to increased numbers of planned CSs (3,4).

    However, the Term Breech Trial was not universally accepted (5,6). In particular, its validity was questioned because the level of care in some centers was not comparable to Western hospitals (7). Due to criticism of the Term Breech Trial, and the fact that many non-randomized population-based studies failed to find a link between poor perinatal outcome and vaginal breech delivery (8–12), new guidelines were published permitting vaginal delivery in selected cases of breech presentation (13).

    In Finland, the vast majority of breech deliveries take place in hospitals with continuous anesthesia and pediatric coverage, as well as equipment for cardiotocographic and ultrasonographic monitoring of the fetus. Moreover, pregnancy care ensures that at-risk pregnancies are identified and the mode of delivery planned accordingly. After the Term Breech Trial, the role of planned CS as the method of delivery has increased also in our hospital, but for a significant proportion of breech presentations a trial of vaginal delivery is still allowed.

    The aim of the present study was to assess selective vaginal breech delivery in the era after the Term Breech Trial. During the study period, the results of the Term Breech Trial had led to an elevated rate of elective cesarean sections, but vaginal delivery still remained an acceptable option. The neonatal and maternal outcomes of planned vaginal breech delivery (planned VD) were compared with those after planned breech cesarean section (planned CS) or planned vaginal delivery in vertex presentation.

    Key Message

    With careful selection of patients and a low trigger point for change in mode of delivery, vaginal breech delivery is still an acceptable option.

    Material and methods

    The patient records of Tampere University Hospital from January 2004 to January 2009 were searched for deliveries with fetuses in breech presentation. As this was a cohort study and deliveries were managed according to the hospital's standard protocol, informed consent of the patients was not required. The study protocol was approved by the ethical committee. Only live singleton fetuses with a gestational age of at least 37 weeks were included. These were divided into two groups, depending on the intended method of delivery. Those that were planned to be delivered by elective cesarean section were classified as planned CS, whereas those planned to be delivered vaginally were classified as planned vaginal delivery, regardless of the actual mode of delivery. Some of the breech presentations were not diagnosed until late pregnancy, some not even before the onset of labor, and these were classified as vaginal if a trial was allowed and as planned CS if not, even though the actual mode of delivery in these cases was mostly emergency CS. As a result of this, some of the babies in the elective CS group were born after a gestational age of 40 weeks. For every planned vaginal breech delivery, a term, singleton fetus in vertex presentation was selected as a control. The vertex control was the next delivery recorded after the breech delivery in the delivery room logbook, excluding elective CSs. Nineteen infants suffering from congenital diseases were excluded, and the final study population totaled 1008 deliveries, of which 497 (66.2% of breech deliveries) were in the planned CS group, 254 (33.8% of breech deliveries) in the planned VD group, and 257 in the vertex group. The delivery unit is a tertiary center with about 5200 annual deliveries. During the study period, there were a total of 24 850 deliveries in our hospital, and the incidence of term, singleton breech delivery was thus 3.1%. The overall CS rate during the study period varied between 15.6 and 18.1%.

    The data were obtained mainly from the mother's medical records concerning the delivery, which also included summary information on the newborn infant's health during the mother's hospital stay. If the child had health problems, the pediatric records were also examined.

    A little less than half (44.7%) of the breech presentations had been subjected to an attempted version. During the study period, 40% of attempts were successful. The mothers willing to attempt vaginal delivery were examined by pelvic X-ray or magnetic resonance imaging to rule out a narrow pelvic outlet (defined as interspinous diameter <10 cm, intertuberous diameter <10 cm, sagittal outlet diameter <10 cm, or the sum of these three diameters <32 cm). The fetus was evaluated by ultrasonography, and if the estimated fetal weight exceeded 4000 g, vaginal delivery was contraindicated. During labor, the fetal heart rate was continuously monitored, and if there were signs of serious fetal distress, an emergency CS was performed (crash emergency CS within 10 min, and in less urgent situations, emergency CS within 20–40 min). Fetuses were delivered by obstetricians or experienced residents. As is the hospital's practice, the Løvset and Mauriceau maneuver was generally used. In our hospital, forceps are not used in breech deliveries. The criteria for breech delivery at our hospital are similar to those presented in the Society of Obstetrics and Gynecology of Canada guidelines in 2009 (14).

    Two previous CSs were considered an indication for an elective CS, as well as hyperextension of the fetal head and any presentation other than frank or complete breech. However, the most common indication was either the mother's fear of vaginal delivery or her expressed desire for an elective CS.

    The primary end-point was neonatal mortality or serious morbidity and serious maternal morbidity, according to the criteria of the Term Breech Trial (1). Low Apgar scores, defined as ≤6, at one or five minutes, and an umbilical artery pH <7.05 were secondary end-points.

    All statistical analyses were performed using SPSS for Windows 14.0 (SPSS Inc., Chicago, IL, USA). Quantitative data were expressed as medians with minimum and maximum values. The results of categorical variables were described by percentages. The comparisons were performed between the planned VD and planned CS groups, and also between the planned VD and vertex groups. The Mann–Whitney U-test, Fisher's exact test and chi-squared test were used, as appropriate. A p-value of <0.05 was considered statistically significant. All p-values are two-sided.

    Results

    The mode of delivery changed significantly more often and was more often emergency CS in the planned VD group compared with both other groups. Of all the breech deliveries, 185 (24.6%) were actually vaginal, 410 (54.6%) elective CSs, and 156 (20.8%) emergency CSs. Figure 1 shows the intended and actual modes of delivery of the study population.

    Details are in the caption following the image

    The intended and actual modes of delivery within the study group. Abbreviations: CS, cesarean section; planned VD, planned vaginal breech delivery; and planned CS, planned breech cesarean section.

    Demographic data and some other maternal and delivery parameters compared between the groups are detailed in Table 1. Mothers in both of the breech presentation groups were significantly more often primiparous than mothers in the vertex group. There was also a significant difference in chronic maternal morbidity between the breech delivery groups. The prevalence of uterine and genital anomalies was higher among the breech groups compared with the vertex group. Breech presentation was diagnosed before the onset of labor significantly more often in the planned CS group. When comparing the breech groups, the pregnancy continued over 41 weeks significantly more often in the planned VD group, and the median duration of pregnancy was slightly longer. In the vertex group, the gestational age at delivery was even more often over 41 weeks and the median duration of pregnancy the longest.

    Table 1. Demographic data and some parameters concerning the mother or the delivery, compared in the groups of planned vaginal breech delivery (planned VD), planned breech cesarean section (planned CS) and vertex delivery.
    Planned VD (n = 254) Planned CS (n= 497) p-Value (planned VD vs. planned CS) Vertex (n = 257) p-Value (planned VD vs. vertex)
    n or median Percentage or range n or median Percentage or range n or median Percentage or range
    Mother's age (years) 29 18–44 30 16–45 0.014 29 17–46 0.67
    Primiparous 168 66.1 346 69.6 0.33 111 43.2 <0.001
    No previous vaginal deliveries 174 68.5 393 79.6 0.001 121 49.0 <0.001
    Gestational diabetes 33 13.0 66 13.3 0.91 31 12.1 0.75
    Pregestational diabetes 1 0.4 5 1.0 0.37 2 0.8 0.57
    Hypertensive conditions 16 6.3 44 8.9 0.22 12 4.7 0.42
    Any chronic illness 30 11.8 112 22.5 <0.001 26 10.1 0.54
    Uterine or genital anomaliesa 6 2.4 20 4.0 0.24 0 0.013
    Attempted cephalic version 120 47.2 216 43.5 0.32
    The breech presentation was not diagnosed before labor 43 16.9 29 5.8 <0.001
    The mode of delivery was not the planned one 80 31.5 103 20.7 0.001 38 14.8 <0.001
    Emergency cesarean section 64 25.2 92 18.5 0.033 13 5.1 <0.001
    Gestational age at delivery weeks 39+6 37+0–42+2 39+1 37+0–42+2 <0.001 40+1 37+0–42+3 0.004
    • Note: Values are expressed as medians and ranges or n and percentages.
    • aMostly unicornic uterus or uterus duplex.

    There was no neonatal mortality in the study population. Newborns in both the planned CS group and vertex group had Apgar scores of 7 or higher more often and lower scores less often than in the planned VD group at the age of one minute, but there was no significant difference at the age of five minutes. Also reflecting the less favorable immediate outcome of the delivery, infants in the planned VD group had an umbilical cord pH of <7.05 significantly more often than infants in either the planned CS or vertex group. Infants in the planned VD group suffered significantly more often from neonatal infections compared with the planned CS group, but the rate of infections in the planned VD group was similar to the vertex group. However, there was no difference between the groups in the frequency of admission to the neonatal intensive care unit (NICU). Neonatal outcome in the groups is shown in Table 2.

    Table 2. Neonatal outcome compared in the groups of planned VD, planned CS and vertex delivery.
    Planned VD (n = 254) Planned CS (n= 497) p-Value (planned VD vs. planned CS) Vertex (n = 257) p-Value (planned VD vs. vertex)
    n or median Percentage or range n or median Percentage or range n or median Percentage or range
    Birthweight (g) 3320 1935–4543 3380 1750–4760 0.33 3630 2320–4775 <0.001
    One minute Apgar
    ≥7 193 76.0 459 92.4 <0.001 246 95.7 <0.001
    <4 15 5.9 8 1.6 0.001 6 2.3 0.042
    Five minute Apgar
    ≥7 250 98.4 493 99.2 0.33 255 99.2 0.40
    <4 0 1 0.2 0.47 1 0.4 0.32
    Umbilical artery pH 7.28 6.80–7.46 7.32 7.01–7.52 <0.001 7.29 7.05–7.54 0.044
    Missing data 6 2.4 27 5.4 0.052 4 1.6 0.51
    Umbilical artery pH <7.05 6 2.4 1 0.2 0.004 0 0.013
    Child admitted to NICU 2 0.8 4 0.8 0.98 1 0.4 0.56
    Perinatal infection 11 4.3 6 1.2 0.006 10 3.9 0.80
    Hyperbilirubinemia 7 2.8 8 1.6 0.29 7 2.7 0.98
    Hypoglycemia 12 4.7 16 3.2 0.30 3 1.2 0.017
    • Note: Values are expressed as medians and ranges or n and percentages. Abbreviation: NICU, neonatal intensive care unit.

    The occurrence of severe morbidity was low. Three neonates among breech infants (one in the planned CS and two in the planned VD groups) and three in the vertex group had serious neonatal morbidity, according to the criteria of the Term Breech Trial (1). There was no difference in overall neonatal morbidity between the groups. There were two birth traumas in the planned VD group, one in the vertex group and none in the planned CS group. In contrast, there were no occurrences of long NICU stay or Apgar scores of <4 in the planned VD group, whereas in the other groups there was one of each. The incidents of severe morbidity, defined according to the Term Breech Trial (1), are detailed in Table 3.

    Table 3. Serious neonatal morbidity, defined according to the Term Breech Trial (1), excluding significant congenital illnesses – case reports.
    Group Actual mode of delivery Description
    Vertex Vaginal delivery, 38+3 weeks Apgar 8–2-8a, pH 7.35, no need for NICU. Pediatric diagnosis: wet lung
    Vertex Vaginal delivery, 38+1 weeks Apgar 9–9b, pH 7.27, no need for NICU. The umbilical cord was ruptured and baby received transfusion of red blood cells
    Vertex Acute cesarean section, 38+2 weeks Apgar 2–5b, pH 7.19, eight days in NICU. Diagnosis: asphyxia
    Intended vaginal Vaginal delivery, 40+0 weeks Apgar 8–9b, pH 7.33, no need for NICU. Diagnosed with mild Erb's palsy and recovered in two weeks
    Intended vaginal Vaginal delivery, 39+2 weeks Apgar 4–7-9a, pH 7.21, no need for NICU. Diagnosis: clavicular fracture
    Planned CS Elective CS, 40+3 weeks Apgar 1–2-8a, pH 7.30, six days in NICU. Diagnosis: meconium aspiration syndrome
    • aApgar score at 1, 5 and 15 minutes of age
    • bApgar score at 1 and 5 minutes of age
    • Abbreviation: NICU, neonatal intensive care unit.

    In the planned CS group, one mother died of a pulmonary embolism 12 days after an elective CS. The operation included hysterectomy due to uncontrolled bleeding because of a placenta previa. Women in the planned CS group suffered significantly more often from massive bleeding (defined as >1500 mL) requiring transfusions. There were no significant differences between the groups in the frequency of puerperal infections or surgical complications (Table 4).

    Table 4. Maternal complications compared in the groups of planned VD, planned CS and vertex delivery.
    Planned VD (n = 254) Planned CS (n= 497) p-Value (planned VD vs. planned CS) Vertex (n = 257) p-Value (planned VD vs. vertex)
    n Percentage n Percentage n Percentage
    Death 0 1 0.2 0
    Bleeding >1500 mL 5 2.0 40 8.0 0.001 6 2.3 0.78
    Received transfusions 8 3.1 33 6.6 0.046 6 2.3 0.57
    Curettage 3 1.2 1 0.2 0.08 3 1.2 0.99
    Evacuated hematoma 0 4 0.8 0.15 0
    Puerperal infectionsa 25 9.8 36 7.2 0.22 17 6.6 0.18
    Ileus 1 0.4 1 0.2 0
    Surgical injury 0 2 0.4 0
    Any complication 40 15.7 77 15.5 0.93 27 10.5 0.08
    • Note: Values are expressed as n and percentages.
    • aExcluding mastitis and respiratory infections.

    After comparing the intention-to-treat groups, some analysis on the breech deliveries was made of the groups according to the actual mode of delivery. Severe bleeding (>1500 mL) was significantly more common among cesarean sections, occurring in 1.1% of vaginal deliveries, 8.0% of elective CSs (p = 0.001 in comparison to vaginal deliveries) and 6.4% of emergency CSs (p = 0.008 in comparison to vaginal deliveries). There were no statistically significant differences in neonatal morbidity or admittance to the NICU.

    When comparing primiparous and multiparous women who actually gave birth vaginally, it was noted that breech infants of primiparous mothers received Apgar scores of <4 at one minute significantly more often (9.3 vs. 0%; p = 0.010) than breech infants of multiparous women, but there were no significant differences in the Apgar scores at five minutes. Median pH was lower (7.22, range 6.89–7.46) in the breech infants born to primiparous women in comparison to breech infants born to multiparous women (7.31, range 7.09–7.46; p < 0.001). Median pH of breech infants born to primiparous mothers was also lower in comparison to vertex infants born to primiparous women (7.29, range 7.09–7.41; p = 0.002). There was no difference in NICU admittance or infant morbidity between the primiparous and multiparous breech delivery groups.

    Discussion

    As in many centers with the tradition of vaginal breech delivery, the guidelines following the publication of the Term Breech Trial were not entirely accepted in our hospital, even though the frequency of elective CS as the method of breech delivery has increased. Our study was a follow-up to the study by Uotila et al. done in 1995–2002 in the same hospital (8), and it was noted that the proportion of intended CSs has increased from one in three to two in three. In the present study, neonatal morbidity remained low and was even lower in both of the breech groups when compared with the previous study (8). The large proportion of primiparous women in the breech groups compared with the vertex group was probably because external version is more often successful among multiparous women. The parity of the vertex group was not matched to the breech group. As there were significantly more primipara in the breech group, the comparison to vertex deliveries might include some bias. However, as the delivery outcome is more favorable in multiparous than in primiparous women, this selection bias might make our study underestimate the already good results. The high proportion of primipara among the breech deliveries also partly explains the fact that the actual manner of delivery was changed from vaginal to CS in 31% of the deliveries in the planned VD group. However, the rate of change in delivery mode was comparable to that of recent studies (15,16). The higher prevalence of maternal chronic illnesses in the planned CS group is probably partly due to the fact that vaginal delivery was sometimes contraindicated because of the mother's illness.

    In accordance with previous reports (1,9,17,18), vaginal breech delivery seemed to predispose infants to low one minute Apgar scores and lower umbilical cord pH in our material, especially in primiparous mothers. Birth traumas, although rare, and perinatal infections were also more common among the babies born vaginally. However, no evidence of overall increased mortality or severe morbidity among infants either born or intended to be born vaginally emerged in the present study, nor were there differences in the five minute Apgar scores or the proportion of neonates admitted to the NICU. Two recent studies have provided similar results and concluded that, if strict selection criteria are met, vaginal delivery is as safe as planned CS and lacks the possible complications in the next pregnancy (16,19). In one single center, the rate of vaginal deliveries has even been increased without impairing the neonatal outcome by using a decision protocol (15). In our study, CS seems to have increased the mother's risk of massive bleeding. The maternal death preceded by a complicated CS and massive bleeding shows that, though usually safe, CS carries a small chance of life-threatening complications. Previous reports have also shown that CS carries an increased risk of immediate as well as long-term maternal complications (20–24). Moreover, it has been suggested that vaginal delivery plays an important role in the maturation of the newborn (25,26). The long-term effects of vaginal delivery were not studied, but previous studies have not found differences in long-term neurological morbidity (9,27).

    Given the overall good neonatal outcome and the increased risk of massive bleeding in the planned CS group, our study demonstrates that the dilemma of breech delivery remains. With a proper selection of patients for trial of vaginal delivery, and a low trigger point for change in mode of delivery if there are signs of asphyxia or poor progress of labor, vaginal breech delivery is still an acceptable option.

    Funding

    No specific funding.