Recent studies have shown the prevalence of persistent occiput posterior position in term, cephalic singletons to be about 5%,1–3 yet few studies have focused on the morbidity associated with this common malposition. Studies have consistently demonstrated an association of persistent occiput posterior position with prolonged labor,2–6 assisted vaginal delivery,1–3,5,6 and the use of epidural analgesia.1,3,6 However, findings regarding the association of fetal occiput posterior position with induction of labor,1,6 cesarean delivery,1,3,5,6 gestational age,1,2,6 and birth weight1–4,6 have been less consistent, with only some studies finding an association.
There are several reasons why the results of these studies may vary. First, the populations in the previous studies differ with regard to inclusion and exclusion criteria. For example, some studies were limited to nulliparous women,1,3 whereas others included both nulliparous and multiparous women.2,4–6 The single study we identified that reported rates of fetal occiput posterior according to parity found a higher rate in nulliparas.6 Because length of labor,7 delivery method (ie, cesarean and assisted vaginal delivery),8 and use of epidural analgesia9 also differ by parity, analyses that do not take parity into account could result in different, less accurate information about the effects of the occiput posterior fetal position on outcome. In addition, some studies included all deliveries,2,5 whereas others limited their study population to women with term, singleton pregnancies.
Differences in outcomes associated with fetal occiput posterior position may also reflect differences in labor management practices, which vary greatly around the world and over time. For example, the recent study from Ireland6 used an active management of labor protocol for nulliparous women, and in the Finnish study from the early 1990s,2 only 3% of patients had epidural analgesia. Only one study (Pearl et al4) has been conducted in the United States, and those data were collected well over a decade ago. In addition, that study did not examine the effect of the occiput posterior position on cesarean and assisted vaginal delivery rates.
We conducted our study to examine labor outcomes in the presence of persistent occiput posterior fetal position, given current labor management practices in the United States.
MATERIALS AND METHODS
Our base population included all women who delivered at Brigham and Women's H4ospital during 1998. The Human Research Committee at the hospital approved the project. Brigham and Women's Hospital is a tertiary care hospital serving a diverse ethnic and cultural population. Data were obtained from an established perinatal database. Data for the database were collected from medical records by well-trained medical record abstractors.
The population for the current analysis was limited to women with term, singleton pregnancies with a vertex fetus, who had spontaneous or induced labor (3315 nulliparas and 3691 multiparas). The occiput posterior position was considered present if it was recorded in the medical record as the position of the fetus at the start of the delivery. Pregnancies complicated by pregestational or gestational diabetes, polyhydramnios, oligohydramnios, intrauterine growth restriction, intrauterine fetal demise, and uterine anomalies were excluded (n = 450) because all of these conditions influence labor management and might also influence the natural position of the fetal head. The proportion of women with fetuses in the occiput posterior position was calculated using all term, laboring women with singleton, vertex fetuses as the denominator. For subsequent analyses, women with a fetus in the occiput transverse position (n = 122, 91 nulliparas and 31 multiparas) were excluded because we were comparing outcomes in women with occiput anterior and occiput posterior fetal positions. After these exclusions, there were 6434 women (2997 nulliparas and 3437 multiparas) included in the final analysis.
Maternal demographic characteristics, obstetric history, and characteristics of the current pregnancy and labor were compared for women with fetuses in the occiput anterior and occiput posterior positions. Labor induction included women induced with either prostaglandin or oxytocin. Prolonged labor was defined as more than 12 hours for the length of the first and second stages combined, more than 10 hours for the first stage, and more than 2 hours for the second stage. Chorioamnionitis was considered present if the provider recorded the diagnosis in the patient's chart. Assisted vaginal deliveries included forceps (including rotations) and vacuum extractions, with approximately one third forceps and two thirds vacuum deliveries. Because the association with occiput posterior did not differ by the type of assisted vaginal delivery, the two were combined in the analysis. Excessive blood loss was defined as more than 500 mL for a vaginal delivery and more than 1000 mL for a cesarean delivery. Postpartum infections included endometritis, wound infections, and “other” infections such as urinary tract infections and pneumonia.
We first examined the entire study population without distinguishing between nulliparas and multiparas. We then examined these parity groups separately to determine whether the associations of interest differed between the two groups. All analyses were performed using SAS 8.0 for Windows (SAS Institute Inc., Cary, NC). The statistical significance of associations was determined using χ2 tests for categoric variables and t tests for continuous variables. A P value of less than .05 was considered significant.
The proportion of women with fetal occiput posterior position at delivery in our population of term, vertex singletons who labored was 5.5%. The proportion differed according to parity and was nearly twice as high among nulliparas (7.2%) as among multiparas (4.0%, P < .001).
Maternal demographic characteristics are shown for all labors (nulliparas and multiparas combined) in Table 1. Women with occiput anterior and occiput posterior fetuses were similar with regard to their age, body mass index, and race. However, women with an occiput posterior fetus were somewhat shorter on average (163.6 cm [64.4 in.] for occiput posterior versus 164.3 cm [64.7 in.] for occiput anterior, P = .03). They also were of higher socioeconomic status, a smaller proportion of occiput posterior patients having Medicaid insurance (11.9% for occiput posterior versus 17.2% for occiput anterior, P = .01).
Obstetric and newborn outcomes for the combined group are shown in Table 2. The rate of induction was somewhat higher in the occiput posterior group (35.8%, versus 31.1% occiput anterior), but this difference did not reach statistical significance (P = .06). Prolonged first (48.3% occiput posterior and 30.3% occiput anterior) and second (53.3% occiput posterior and 18.1% occiput anterior) stages of labor, oxytocin augmentation (48.9% occiput posterior and 36.8% occiput anterior), use of epidural analgesia (86.1% occiput posterior and 73.1% occiput anterior), chorioamnionitis (4.7% occiput posterior and 1.1% occiput anterior), assisted vaginal delivery (24.6% occiput posterior and 9.4% occiput anterior), cesarean delivery (37.7% occiput posterior and 6.6% occiput anterior), and third or fourth degree laceration (18.2% occiput posterior and 6.7% occiput anterior) were all strongly associated with occiput posterior delivery (P < .001). Excessive blood loss was more likely in the occiput posterior group (13.6% versus 9.9% occiput anterior, P = .03), as was postpartum infection (2.2% occiput posterior and 0.8% occiput anterior, P = .01). Within postpartum infection, wound infection was significantly more common with the occiput posterior position (0.8%, versus 0.1% occiput anterior, P = .01), but endometritis was not (0.6% occiput posterior versus 0.3% occiput anterior, P = .4).
Infants delivered from occiput anterior and occiput posterior positions had similar gestational ages and birth weights. Interestingly, there was a higher proportion of male infants in the occiput posterior group (56.4%, versus 51.1% occiput anterior, P = .048). One-minute Apgar scores less than 7 were more common in the occiput posterior group (12.4%, versus 7.1% in occiput anterior, P = .001), but low 5-minute Apgar scores were not (0.6% occiput posterior and 0.9% in occiput anterior, P = .5). Shoulder dystocia and nuchal cord also did not differ in the occiput posterior and occiput anterior groups.
Table 3 displays maternal demographic characteristics for nulliparas and multiparas separately. Several significant differences in the factors associated with occiput posterior position were found between these two groups. Nulliparous women with occiput posterior fetuses were shorter on average than women with occiput anterior fetuses, but this was not true among multiparous women. In addition, among nulliparas a higher proportion of women in the occiput posterior group were black and a smaller proportion of women were white, although this was not true for multiparas. Occiput posterior position was associated with older maternal age in the multiparous group of women but not in the nulliparous group. Also, a difference in socioeconomic status was significant in the overall comparison but not in the separate comparisons within parity groups. However, in this case, the direction of the association was the same in nulliparas and multiparas (a lower proportion of patients with Medicaid insurance in the occiput posterior groups). Finally, in the multiparous group of women a history of a cesarean delivery was more common in women with the occiput posterior position (21.9%, versus 14.4% occiput anterior, P = .02).
Obstetric and newborn outcomes are listed separately for nulliparas and multiparas in Table 4. Many of the differences in obstetric outcome noted in the combined group remained significant in the parity subgroups, including longer labor, use of epidural analgesia, assisted vaginal or cesarean delivery, and third or fourth degree laceration. A second stage of labor longer than 2 hours was strongly associated with the occiput posterior position in both nulliparas and multiparas. Similarly, operative (vaginal and cesarean) delivery also remained strongly associated with the occiput posterior position in both nulliparas and multiparas. However, several of the differences between the occiput anterior and occiput posterior deliveries were no longer significant in the multiparous group when examined separately. These outcomes included prolonged first stage of labor, oxytocin augmentation, chorioamnionitis, excessive blood loss, and postpartum infection. For all these variables except oxytocin augmentation, the direction of the association in the multiparous group was the same as in the combined and nulliparous groups.
As in the combined group, gestational age and birth weight were similar in the occiput anterior and occiput posterior positions in both nulliparous and multiparous women. Although male fetuses had a higher rate of the occiput posterior position in both nulliparas and multiparas, it no longer reached significance in either group. Only the nulliparas were significantly more likely to have a 1-minute Apgar score less than 7 after occiput posterior delivery. Among multiparas, a higher proportion of women with occiput posterior fetuses had lower 1-minute Apgar scores, but the association did not reach statistical significance. The 5-minute Apgar, a better predictor of neonatal morbidity, remained no different in the occiput posterior as compared with the occiput anterior in nulliparas and multiparas. As in the combined group, no difference was seen in shoulder dystocia or nuchal cord with the occiput posterior position in nulliparas and multiparas.
Our study confirms that persistent fetal occiput posterior position occurs in approximately 5% of singleton, vertex, term labors and is more common in nulliparas than multiparas. It also supports the reported association of the occiput posterior position with slower progress of labor and a higher rate of cesarean and operative vaginal deliveries.1–6 In addition, we identified a higher rate of a number of complications including chorioamnionitis, third and fourth degree lacerations, excessive blood loss, and postpartum infection.
The impact of persistent occiput posterior position on method of delivery in a nulliparous patient in our population was so strong as to reduce the chance of a spontaneous vaginal delivery to only 26%. Multiparas fared somewhat better but still had only a 57% chance of having a spontaneous vaginal delivery. This low rate of spontaneous vaginal delivery among nulliparas with persistent occiput position in our study is remarkably consistent with the 27% spontaneous vaginal delivery rate reported by Floberg et al in a population of nulliparas with the spontaneous onset of labor.3 The findings of Fitzpatrick et al6 were also very similar to ours, with a 29% spontaneous vaginal delivery rate for nulliparas and a 55% rate for multiparas.
Our finding of a higher rate of intrapartum and postpartum complications such as third and fourth degree lacerations, excessive blood loss, and postpartum infection is likely to reflect the higher proportion of women with cesarean and assisted vaginal deliveries. Assisted vaginal deliveries are known to be associated with a higher rate of third and fourth degree lacerations.10 A higher rate of postpartum hemorrhage with assisted vaginal delivery has also been reported.11 The increase in wound infection in the occiput posterior group is most likely due to the higher proportion of women with wounds from perineal lacerations and cesarean deliveries. We also noted a higher rate of “chorioamnionitis” as diagnosed by physicians. Although this could reflect an increase in infection, it is more likely due to the higher proportion of women in the occiput posterior group who received epidural analgesia. Previous studies have demonstrated that most fever in term labor is associated with the use of epidural analgesia and likely to be noninfectious in origin.12
Our finding of a higher rate of epidural use in women with the fetal occiput posterior position has been documented in other studies.1,3,6 However, the reason for the association is not clear. It is not known whether epidural analgesia may contribute to persistent occiput posterior position, or whether women with the occiput posterior position have more painful labors and request epidural analgesia more often. A higher rate of fetal malposition (occiput posterior or transverse) with epidural analgesia was identified in one small randomized trial,13 but the data are not sufficient for a definitive conclusion. Thus, although this study has found an association with epidural analgesia, it does not suggest that epidural analgesia causes the occiput posterior position. Further study would be needed to clarify this issue.
We found a somewhat higher rate of induced labor in our occiput posterior population, though the association did not reach statistical significance. Only Fitzpatrick et al6 identified significantly more inductions in their nulliparous occiput posterior group. Neri et al,5 in fact, found fewer inductions in their occiput posterior group. Taken together, these data suggest that there is at most a modest association of induction with the occiput posterior position.
We also found an association between prior cesarean delivery and the occiput posterior position in multiparas. Neri et al5 found the same trend in their population, though the difference did not reach statistical significance.
Some of our findings were unexpected, such as the association of higher socioeconomic status with the occiput posterior position, the greater frequency of black nulliparas in the occiput posterior group, and the association of older maternal age in multiparas with persistent occiput posterior position. Because we know of no mechanism that would explain these associations, we have to consider that some of them may have occurred in our study by chance. Additional studies are needed to draw any conclusions about these associations.
Lastly, we looked at neonatal outcomes. We found significantly lower 1-minute Apgar scores among occiput posterior infants in the combined and nulliparous groups. Floberg et al3 found the same in their nulliparous population. This may reflect the longer second stage seen in nulliparas with an occiput posterior position, which was nearly 3 hours in our study. As in previous studies, we found no difference in 5-minute Apgar scores between infants born in the occiput anterior and occiput posterior positions. We also found some excess of male fetuses in the occiput posterior group. This could be related to the slightly larger head circumference in male infants.14
One limitation of our study is the minimal information in our perinatal database on neonatal outcome, so that we were unable to comment on neonatal outcome beyond the 5-minute Apgar score. We were not able to confirm or refute the findings of an increased rate of Erb and facial nerve palsies reported by Pearl et al.4 In addition, because our data were obtained retrospectively from medical records, we were unable to verify the accuracy with which delivery position was recorded. Finally, some of the associations that were significant in the overall group were no longer significant when the data were stratified by parity. In many cases, the direction of the association was the same, and the lack of statistical significance may reflect the smaller numbers in the stratified groups.
In summary, our data confirm that persistent occiput posterior position is a common malposition in labor that is associated with a significantly lower rate of spontaneous vaginal delivery and often requires an assisted delivery to effect a vaginal birth, or a cesarean delivery. It is associated with a variety of other adverse maternal outcomes such as prolonged labor, chorioamnionitis, third or fourth degree perineal laceration, excessive blood loss, and postpartum infection. In addition, occiput posterior newborns are more likely to have a low 1-minute Apgar score. Further studies to evaluate the causes and consequences of fetal occiput posterior position are warranted.