Historically X-ray studies have demonstrated that the fetal head typically engages in the transverse diameter late in the third trimester1 and usually rotates in labor to an occipito-anterior (OA) or occipito-posterior (OP) position. However, OP position is thought to occur in 15–20% of women before labor at term.1,2 Standard teaching is that 90–95% of these rotate during labor once the head reaches the pelvic floor3 and that most OP positions at delivery arise due to a persistent OP position.4 The incidence of OP position at delivery has been shown to be between 1–5%2,5–10 and has been associated with many factors and outcome measures in the pregnancy and delivery, including an increased cesarean delivery rate.6,8,9,11,12
Traditionally fetal head position is determined by clinical palpation of the abdomen (Leopold’s maneuvers) and vaginal examination of the fetal head. There are several problems with these methods. What little data there is on clinical palpation assesses the position of the fetal spine, but this does not always correspond to the fetal head position.13,14 Once the cervix has dilated in labor, vaginal examination is possible, but this may be difficult in early labor and is highly subjective.
Transabdominal ultrasound examination has been used in labor and before instrumental delivery to confirm fetal head position. It has been shown to be easy to perform and highly reproducible.15 It has demonstrated the high error rate of vaginal examination in determining fetal head position in the first16,17 and second stages18–20 of labor and before instrumental delivery.21 There is little data on the incidence of sonographic OP position before induction of labor, but one group found a high incidence of 60%.22 Its use before induction of labor to predict delivery outcome has shown conflicting results.23,24 The aim of this study was to assess clinical and sonographic fetal head position before induction of labor, to compare this to head position at delivery, and to assess whether fetal head position before induction of labor predicted obstetric outcome.
MATERIALS AND METHODS
We conducted a prospective observational study between June 2001 and November 2003, during which we recruited 289 women undergoing induction of labor at University College Hospital, London, UK. All recruited women presented with pregnancies greater than 35+6 weeks of gestation with a cephalic singleton fetus. Women with an intrauterine fetal death or an antepartum hemorrhage as the indication for induction were excluded. Ethical approval was obtained from the Trust Ethics Committee, and written informed consent was obtained from the women before the procedure.
Data were obtained from the maternal notes on maternal age, ethnicity, body mass index, and parity. Gestational age was calculated by using the last menstrual period combined with a first trimester scan. Abdominal palpation was performed on all women, and Leopold’s maneuvers were used to determine fetal position. A transabdominal ultrasound scan was then performed with an Aloka 1700 ultrasound machine (Aloka Company Ltd, Mitaka-shi, Tokyo, Japan). Fetal head position, fetal spine position, and placental position were assessed with a 3.5-MHz probe. The fetal head position was obtained by placing the transducer suprapubically in the transverse plane and visualizing the fetal orbits and midline cerebral echo in a transverse view of the head. The fetal spine position was obtained by placing the transducer along the sagittal plane of the thoracic spine. Both fetal head and spine positions were defined as direct occipito-anterior, right occipito-anterior, left occipito-anterior, right occipito-transverse, left occipito-transverse, direct occipito-posterior, right occipito-posterior, or left occipito-posterior. If the operator was unable to determine the fetal position, it was classed as “undefined.” All examinations and scans were performed immediately before labor induction by the same operator (E.P.), who was experienced in the use of ultrasonography. The women and clinicians were blinded to the ultrasound findings.
The labor induction was then commenced, with the method depending on the Bishop score in accordance with the hospital protocol and The Royal College of Obstetricians and Gynaecologists’ guidelines.25 In a nulliparous woman with a Bishop score of less than four, 2 mg of prostaglandin estradiol gel was inserted in the posterior fornix of the vagina. If the Bishop score was greater than six and the cervix was favorable for artificial rupture of membranes, then this was performed. All other women received 1 mg of vaginal prostaglandin estradiol gel. Examinations during the induction were performed every 6 hours, and artificial rupture of membranes was performed once possible. In accordance with national guidelines, oxytocin infusion was used if labor did not commence after rupture of membranes.25 Once in labor the women were managed according to the hospital protocol and decisions made by the attending clinicians, with vaginal examinations every 4 hours and the use of oxytocin infusion for slow progress according to the Partogram. Cesarean delivery was performed as necessary. The outcomes of labor and delivery were obtained from the women’s notes after delivery. Using the same criteria as the ultrasound position, fetal head position at delivery was defined as head position at spontaneous vaginal delivery or vaginal assessment of position before instrumental delivery or at cesarean delivery, as noted by the midwife or doctor at delivery, as is standard practice in our unit. Length of labor was defined as the time from the onset of regular contractions until delivery.
The primary outcome was mode of delivery, and fetal head position at delivery and length of labor were secondary outcomes. Assuming a cesarean delivery prevalence of 30% (the cesarean rate for induction of labor in the unit at the time), it was estimated that the sample size of 289 would give an 80% probability of detecting a clinically significant increase in the cesarean delivery rate from 30% to 45%. All statistical analyses were carried out with SPSS 10.1 for Windows (SPSS Inc, Chicago, IL). Statistical significance between the groups for the demographic data were determined by the χ2 test for categorical variables and an independent samples t test or Mann-Whitney U test for continuous variables. Maternal age, gestational age, and body mass index were examined as continuous variables, and parity and ethnicity as categorical variables. The χ2, Mann-Whitney U, t test, and analysis of variance (ANOVA) were used to assess the significance of the differences between OP and non-OP positions for each of the outcomes. Chi-square was used for the categorical outcome measures and the Mann-Whitney U, t test, or ANOVA for the continuous variables. Logistic regression was used to determine which of the predictors (maternal age, gestational age, body mass index, parity, ethnicity, and OP position before induction of labor) were independent predictors of cesarean delivery. P<.05 was considered statistically significant.
All 289 women recruited to the study had an abdominal palpation and ultrasound scan performed before commencement of induction of labor. There were 175 nulliparous and 114 multiparous women, with a median age of 32 years (range 17–45). The median body mass index was 23 (range 17–43), and there were 238 white and 51 nonwhite women. The median gestational age at induction of labor was 41 weeks (range 36–42). The indications for induction of labor in the 289 women were 172 (59%) for postdates, 20 (7%) for hypertensive disease, 18 (6%) for preterm or term prelabor rupture of the membranes, 14 (5%) for maternal request (nonmedical reasons), 14 (5%) for intrauterine growth restriction and/or reduced amniotic fluid volume, 9 (3%) for poor obstetric history, 8 (3%) for fetal abnormality, 8 (3%) for a clinically large baby and/or hydramnios, 8 (3%) for obstetric cholestasis, 8 (3%) for prepregnancy or gestational diabetes, and 10 (3%) for another maternal medical disease or obstetric reason.
The results of abdominal palpation of spine position and ultrasound examination of head and spine position and position of head at delivery are shown in Table 1. The fetal head was in an OA position in 79 (27%) women, occipito-transverse (OT) position in 110 (38%) women, OP position in 99 (34%) women, and undefined in one (1%) woman on ultrasound examination before induction of labor. Fetal head position was determined on ultrasound examination in 288 of 289 women, and all further analysis regarding fetal position on ultrasonography was performed on these 288 women. Demographic data on these 288 women, according to ultrasound fetal head position (OP versus non-OP), is shown in Table 2. The only significant difference between the two groups was that women who had an OP position on ultrasonography before induction of labor had a slightly lower median age than women with a non-OP position (30 versus 33 years). Data on the Bishop score before induction of labor was available in 283 of the 288 women, and there was no difference between the Bishop score in an OP and a non-OP position before induction of labor when compared with the Mann-Whitney U test (Z=–0.16, P=.88).
Ninety-nine (34%) of the 288 women had an OP position of the fetal head on ultrasonography before induction of labor. Outcome data on fetal head position at delivery was missing in 17 women, and one woman had a fetus that turned to breech presentation before delivery. Therefore, further analysis regarding head position at delivery was performed on the remaining 270 women. Of these 270 women, 97 (36%) had an OP position on ultrasonography before induction of labor. At delivery, eight (8%) of these 97 were OP, 12 (12%) were OT, and 77 (79%) were OA.
Twenty-five babies were in an OP position at delivery. Before induction of labor, eight (32%) of these 25 women were OP, eight (32%) were OT, and nine (36%) were OA. Therefore, 68% of OP positions at delivery occurred due to a mal-rotation from a non-OP position during labor.
Obstetric outcome data were available for all 288 women (Table 3). This table also compares outcome data from those with an OP and those with a non-OP position on ultrasonography before induction of labor. There was no significant difference between the two groups for delivery within 24 hours, rate of vaginal delivery, vaginal delivery within 24 hours, instrumental delivery, abnormal cardiotocograph, use of oxytocin, use of epidural, birth weight, and length of time from the start of induction of labor to delivery. The total length of labor was significantly longer with an OP position before induction of labor compared with a non-OP position (median 606 versus 499 minutes). The length of labor and time from the start of induction of labor to delivery were also compared in the OP and non-OP groups using an analysis of variance (ANOVA) test. There was no evidence for heterogeneity of variance with either comparison. The length of labor was significantly longer in the OP group (F=5.83, P=.016), but there was no significant difference in the total time from the start of induction of labor to delivery (F=1.14, P=.287). Of the 25 infants in OP position at delivery, 24 (96%) were delivered by cesarean delivery and one (4%) by ventouse. Of these 24 women who had a cesarean delivery, two women had a failed ventouse before cesarean delivery, one had a cesarean delivery at full dilatation due to a prolonged second stage and a high head, and 21 had a cesarean delivery before full dilatation (nine for fetal distress, eight for failure to progress, and four for a failed induction of labor–failure to reach 4 cm). Logistic regression demonstrated that only parity and body mass index were independent predictors of cesarean delivery (OR 10.61, 95% CI 4.97–22.64, P=.000 and OR 1.11, 95% CI 1.11–1.26, P=.000, respectively). Occipito-posterior position before induction of labor did not independently predict the risk of cesarean delivery (OR 1.75, 95% CI 0.97–3.15, P=.06).
The data were then analyzed separately according to parity. Before labor, 69 (40%) of the 174 nulliparous women had an OP position on ultrasonography compared with 30 (26%) of the 114 multiparous women. Complete data on fetal head position at delivery were available in 159 nulliparous women and 111 multiparous women. Twenty-one (13%) of the nulliparous women had an OP position at delivery compared with four (4%) of multiparous women. Multiparous women were significantly less likely to have an OP fetal head position both on ultrasonography before induction of labor (χ2=5.43; P=.02) and at delivery (χ2=7.18, P<.01). The obstetric outcomes for the 288 women according to parity and ultrasound fetal head position are shown in Table 4. When obstetric outcomes for nulliparous and multiparous women were analyzed separately, the only significant difference between the two groups was that multiparous women with an OP position before induction of labor were more likely to deliver vaginally than multiparous women with a non-OP position. Overall, the risk of cesarean delivery was significantly higher in nulliparous compared with multiparous women (45% versus 11%, χ2=56.09; P<.01). This difference between nulliparous and multiparous women remained significant even when they were separated into OP and non-OP position groups before induction of labor (39% versus 0%, χ2=16.14, P<.01, and 49% versus 15%, χ2=22.82, P<.01, respectively).
Placental position on ultrasonography was compared with both fetal head and spine position on ultrasonography before induction of labor using a χ2 test. No significant association was found between fetal head position on ultrasonography and anterior (χ2=0.48, P=.49) or posterior placental position (χ2=1.14, P=.29) or fetal spine position on ultrasonography and anterior (χ2=0.93, P=.34) or posterior placental position (χ2=0.76, P=.39).
Of the 267 women in whom a fetal position was defined on abdominal palpation, the spine position was assessed correctly (in relation to the ultrasonographic spine position) in only 92 women (34%). Palpation was within 45° of the correct position in a total of 165 (62%) women, within 90° in 217 (81%), and more than 90° different in 50 (19%) women. The palpation was more likely to be correct when the fetus was lying with its spine in an occipito-anterior position (72 of 128, 56%) compared with non-OA positions (20 of 146, 14%). Of the 288 women where complete ultrasound data were available, the head and spine positions correlated exactly in 137 (48%) women, were within 45° in 274 (95%), and within 90° in 288 (100%).
Our ultrasound data show that fewer than 10% of OP positions before induction of labor remain OP at delivery and that two thirds of OP positions at delivery arise from a mal-rotation from an initial non-OP position. This holds true in both nulliparous and multiparous women. Our findings challenge conventional wisdom regarding the movements of the fetal head in labor.
The incidence of OP position before induction of labor in our study was 34%. One other study has agreed with our data of a higher incidence of OP position before induction of labor,22 compared with the reported incidence before spontaneous labor. One hypothesis to explain this finding is that OP position may in some way delay the spontaneous onset of labor, making induction of labor more likely.
Our findings indicate that 92% of OP positions before induction of labor rotated to an OA position and that 68% of OP positions at delivery developed through a mal-rotation during labor from an initially OA or OT position. This is contrary to standard teaching and other studies in spontaneous labor that state that failure of rotation occurs in up to 20% of cases, resulting in a persistent OP position and that most OP positions at delivery arise from a persistent OP position.3,4,11,26 Some of this variation in findings may be due to differences in the populations studied and the timing of assessment of position, either before or during labor or that the head movements in induced labor are inherently different from those after spontaneous labor. Gardberg and colleagues2 found similar results to ours, demonstrating that OP position at delivery develops more frequently (62%) from mal-rotation from a non-OP position than from failure of rotation from an initial OP position. Their study was, however, in a group of 408 women before both spontaneous and induced labor, rather than a solely induced labor group, supporting the fact that our findings could be extrapolated to a population of spontaneously laboring women.
Occipito-posterior position at delivery has been associated with many factors and outcome measures in the pregnancy and delivery, including short maternal stature,8 nulliparity,8 anterior placenta,27 induction of labor,9 variable fetal heart rate decelerations,5,28 use of epidural analgesia,6,8,9 use of oxytocin augmentation,6,8,9,11 increased length of labor,8,9 increased length of the second stage,10 increased instrumental delivery rate,6,8–10 increased cesarean delivery rate,6,8,9,11,12 increased episiotomy rate,10 perineal trauma,8,9,29,30 larger birth weights,6 and lower 1-minute Apgar scores.5,6,8,9,31 In our study, OP position before induction of labor was associated with a significantly longer labor than a non-OP position. However this finding is not likely to be clinically significant given the wide ranges of the length of labor and that there was no difference in the time from the start of the induction to the delivery. It has been proposed that the increased length of labor associated with OP position may be due to decreased uterine contractility,9 but the lack of association with oxytocin use in our study does not support this.
We did not find that OP position before induction of labor increased the risk of cesarean delivery or was associated with other adverse obstetric outcomes. Most groups have studied the associations of OP position at delivery rather than position before labor. It is only knowledge of position before or during labor that will be potentially clinically useful for prediction of successful outcome and, therefore, appropriate management.
We found that 96% of those with OP position at delivery had a cesarean delivery, but few conclusions can be drawn from these data because we did not examine fetal head position during labor or in the second stage. This, therefore, represents a group of women who are at very high risk of surgical delivery in whom either the head was in an OP position at a cesarean delivery in the first stage or at spontaneous vaginal delivery or in an OP position in the second stage immediately before a trial of instrumental delivery. The majority (84%) of these women were not deliverable vaginally because the cervix was not fully dilated.
We found that there was no difference in the incidence of OP position and non-OP position in white and nonwhite women before induction of labor, which is in keeping with findings in early labor.12 We found that OP position before induction of labor and at delivery is more common in nulliparous than multiparous women. This has been previously noted,8 but the reasons are poorly understood. Occipito-posterior position has been associated with reduced uterine contractility caused by other factors, such as epidural6 and the resulting interventions.9 It could be proposed, therefore, that inefficient uterine activity in labor itself in nulliparity predisposes to the persistent OP position. Our findings, however, do not support this theory; oxytocin requirement was not more common in the OP position group.
In all groups (all women, OP position and non-OP position), cesarean delivery was more common in nulliparous than multiparous women. This reflects the known association between nulliparity and cesarean delivery, especially after induction of labor,24 which appears to be irrespective of fetal position. The finding that multiparous women with an OP position are more likely to deliver vaginally than those with a non-OP position is likely to be due to the small numbers of multiparous women with an OP position and the low risk of cesarean delivery in multiparous women as a whole. This is supported by our finding that no other outcomes were significantly different.
This study has demonstrated the ease of use of transabdominal ultrasonography in determining fetal head position before induction of labor. Previous studies have shown that sonographic head position can reveal the high error rate of vaginal examination in determining fetal head position in the first16,17 and second stages of labor18–20 and that digital vaginal examination is more difficult17, 26 and less accurate16–18,21,26,32 in an OP than a non-OP position. Transperineal ultrasonography has also been used successfully to determine fetal head position when the head is low in the pelvis.20 However, there are few data on its use before induction of labor or comparing sonographic head position before induction of labor with head position at delivery. Previous studies before induction of labor have mainly included one element of a preinduction assessment: one study found that ultrasonographic fetal head position can be used to predict the risk of cesarean delivery and the induction-to-delivery interval.23 However, this has not been confirmed in other studies.24
Our study has shown that abdominal palpation is highly inaccurate in determining fetal head position, particularly in a non-OP position; the spine and head position correlated in only 48% of women. This lack of correlation in orientation of the fetal head and spine has been noted before,13,14 but it is often not clinically appreciated during abdominal palpation. If we assume that it is head position rather than spine position that will determine the position changes in labor, we can conclude that abdominal palpation of fetal position is not a clinically useful procedure. Vaginal determination of head position is only possible once the cervix has dilated, may be subjective, and was not assessed by this study. Determining fetal head position by ultrasonography can confirm position before labor or during labor when the clinician is unsure on vaginal examination. It has particular potential before instrumental delivery and in training.
These findings challenge traditional teaching about the diagnosis of and implications of an OP position at term. The widespread belief that OP position before induced (and spontaneous) labor is associated with poor obstetric outcome does not appear to be well founded. At present, women who are suspected of having a fetus in an OP position are often frightened by this finding and resort to various measures to try to change the fetal position.33 Our data suggest that they can be reassured that, although an OP position at this stage may be associated with a longer labor, it is not associated with an increased risk of cesarean delivery. This is likely to be due to the fact that fetal head position before induction of labor does not predict position late in labor or at delivery.
1. Caldwell WE, Moloy HC, D’Esopo DA. A roentgenologic study of the mechanism of engagement of the fetal head. Am J Obstet Gynecol 1934;28:824–41.
2. Gardberg M, Laakkonen E, Sälevaara M. Intrapartum sonography and persistent occiput posterior position: a study of 408 deliveries. Obstet Gynecol 1998;91:746–9.
3. Cunningham FG, Leveno KJ, Bloom SL, Heath JC, Gilstrap LC, Wenstrom KD. Williams obstetrics. 22nd ed. New York (NY): McGraw Hill; 2005.
4. Myerscough PR. Munro Kerr’s operative obstetrics. 10th ed. London: Baillière Tindall; 1982.
5. Ingemarsson E, Ingemarsson I, Solum T, Westgren M. Influence of occiput posterior position on the fetal heart rate pattern. Obstet Gynecol 1980;55:301–4.
6. Sizer AR, Nirmal DM. Occipitoposterior position: associated factors and obstetric outcome in nulliparas. Obstet Gynecol 2000;96:749–52.
7. Phillips RD, Freeman M. The management of persistent occiput posterior position: a review of 552 consecutive cases. Obstet Gynecol 1974;43:171–7.
8. Ponkey SE, Cohen AP, Heffner LJ, Lieberman E. Persistent fetal occiput posterior position: obstetric outcomes. Obstet Gynecol 2003;101:915–20.
9. Fitzpatrick M, McQuillan K, O’Herlihy C. Influence of persistent occiput posterior position on delivery outcome. Obstet Gynecol 2001;98:1027–31.
10. Neri A, Kaplan B, Rabinerson D, Sulkes J, Ovadia J. The management of persistent occipito-posterior position. Clin Exp Obstet Gynecol 1995;22:126–31.
11. Akmal S, Tsoi E, Howard R, Osei E, Nicolaides KH. Investigation of occiput posterior delivery by intrapartum sonography. Ultrasound Obstet Gynecol 2004;24:425–8.
12. Akmal S, Kametas N, Tsoi E, Howard R, Nicolaides KH. Ultrasonographic occiput position in early labour in the prediction of caesarean section. BJOG 2004;111:532–6.
13. Ververs IA, de Vries JI, van Geijn HP, Hopkins B. Prenatal head position from 12–38 weeks. I. Developmental aspects. Early Hum Dev 1994;39:83–91.
14. Ververs IA, de Vries JI, van Geijn HP, Hopkins B. Prenatal head position from 12–38 weeks. II. The effects of fetal orientation and placental localization. Early Hum Dev 1994;39:93–100.
15. Akmal S, Tsoi E, Nicolaides KH. Intrapartum sonography to determine fetal occipital position: interobserver agreement. Ultrasound Obstet Gynecol 2004;24:421–4.
16. Sherer DM, Miodovnik M, Bradley KS, Langer O. Intrapartum fetal head position. I. Comparison between transvaginal digital examination and transabdominal ultrasound assessment during the active stage of labor. Ultrasound Obstet Gynecol 2002;19:258–63.
17. Akmal S, Tsoi E, Kametas N, Howard R, Nicolaides KH. Intrapartum sonography to determine fetal head position. J Matern Fetal Neonatal Med 2002;12:172–7.
18. Sherer DM, Miodovnik M, Bradley KS, Langer O. Intrapartum fetal head position. II. Comparison between transvaginal digital examination and transabdominal ultrasound assessment during the second stage of labor. Ultrasound Obstet Gynecol 2002;19:264–8.
19. Kreiser D, Schiff E, Lipitz S, Kayam Z, Avraham A, Achiron R. Determination of fetal occiput position by ultrasound during the second stage of labor. J Matern Fetal Med 2001;10:283–6.
20. Chou MR, Kreiser D, Taslimi MM, Druzin ML, El Sayed YY. Vaginal versus ultrasound examination of fetal occiput position during the second stage of labor. Am J Obstet Gynecol 2004;191:521–4.
21. Akmal S, Kametas N, Tsoi E, Hargreaves C, Nicolaides KH. Comparison of transvaginal digital examination with intrapartum sonography to determine fetal head position before instrumental delivery. Ultrasound Obstet Gynecol 2003;21:437–40.
22. Tribalat S, Lewin D, Tochon C. Orientation of the fetal head at the time of engagement in induced labor [in French]. J Gynecol Obstet Biol Reprod 1986;15:937–9.
23. Rane SM, Guirgis RR, Higgins B, Nicolaides KH. The value of ultrasound in the prediction of successful induction of labor. Ultrasound Obstet Gynecol 2004;24:538–49.
24. Peregrine E, O’Brien P, Jauniaux E. Clinical and ultrasound parameters to predict the risk of cesarean delivery after induction of labor. Obstet Gynecol 2006;107:227–33.
26. Souka AP, Haritos T, Basayiannis K, Noikokyri N, Antsaklis A. Intrapartum ultrasound for the examination of the fetal head position in normal and obstructed labor. J Matern Fetal Neonatal Med 2003;13:59–63.
27. Gardberg M, Tuppurainen M. Anterior placental location predisposes for occiput posterior presentation near term. Acta Obstet Gynecol Scand 1994;73:151–2.
28. Sokol RJ, Roux JF, McCarthy S. Computer diagnosis of labor progression. Am J Obstet Gynecol 1975;122:253–60.
29. Pearl ML, Roberts JM, Laros RK, Hurd WW. Vaginal delivery from the persistent occiput posterior position: influence on maternal and neonatal morbidity. J Reprod Med 1993;38:955–61.
30. Sultan AH, Kamm MA, Hudson CN, Bartram CI. Third degree obstetric anal sphincter tears: risk factors and outcome of primary repair. BMJ 1994;308:887–91.
31. Gardberg M, Tuppurainen M. Persistent occiput posterior presentation: a clinical problem. Acta Obstet Gynecol Scand 1994;73:45–7.
32. Rayburn WF, Siemers KH, Legino LJ, Nabity MR, Anderson JC, Patil KD. Dystocia in late labor: determining fetal position by clinical and ultrasonic techniques. Am J Perinatol 1989;6:316–9.